Showing 1–50 of 726 results for author: Chakraborty, S

MINT: Min-Selection Preference Distillation for Balanced Multi-Objective Alignment

Authors: Tony Tu, Sayan Chakraborty, Ruomeng Xu, Tony Qin, Austin Tian

Abstract: Aligning a language agent to several objectives at once is a persistent failure mode of preference-based training: when objectives are combined additively, optimization collapses onto whichever is cheapest to improve and sacrifices the rest, so a support agent learns to sound warm while giving no real help. The root issue is that an additive reward has no notion of balance. We introduce Mint (MIN-… ▽ More Aligning a language agent to several objectives at once is a persistent failure mode of preference-based training: when objectives are combined additively, optimization collapses onto whichever is cheapest to improve and sacrifices the rest, so a support agent learns to sound warm while giving no real help. The root issue is that an additive reward has no notion of balance. We introduce Mint (MIN-selection preference disTillation), a one-line change to preference distillation: rather than ranking sampled candidates by a weighted sum of rewards, we rank them by their weakest objective, distilling the best-balanced candidate over the most lopsided one with an unchanged DPO objective. This is the p -> negative infinity limit of a generalized-mean family spanning additive to worst-case selection. Across cooperative emotional support and adversarial negotiation, min-selection lifts both objectives while sharply cutting their imbalance; on emotional support it raises the weaker axis from 0.37 to 0.64 (p < 10^-40), surpassing human experts and persisting across full multi-turn rollouts. A turn-by-turn analysis yields our central finding: min-selection corrects imbalance in proportion to how imbalanced the reference policy is, and its benefit endures over an interaction precisely as long as that imbalance does. △ Less

Submitted 14 August, 2026; originally announced August 2026.

Unlocking Fractional Moments in Delphic Set Streams

Authors: Aranya Kumar Bal, Sourav Chakraborty, Arijit Ghosh, Rudrayan Kundu

Abstract: We consider estimation of non-integer frequency moments $F_k$ and related Bernstein-type statistics in the Delphic set stream model under a bounded-frequency assumption: every universe element appears at most $τ$ times. The main challenge of this model is to keep space low while also keeping update time low, which is not trivial because the sets can be exponential in size compared to their represe… ▽ More We consider estimation of non-integer frequency moments $F_k$ and related Bernstein-type statistics in the Delphic set stream model under a bounded-frequency assumption: every universe element appears at most $τ$ times. The main challenge of this model is to keep space low while also keeping update time low, which is not trivial because the sets can be exponential in size compared to their representations. Our core insight is that by sampling the stream at different rates and observing the resulting distinct-counts, we can 'probe' the frequency distribution and numerically integrate these probes to reconstruct a broad class of statistics. Building on that, we crucially observe that the distinct-count of a randomly sampled substream, viewed as a function of the sampling rate, is a single analytic object whose evaluations determine a broad class of statistics via a complementary Laplace-type integral. Algorithmically we exploit this by: 1. estimating those evaluations using only standard $F_0$ (distinct-count) algorithms on sampled substreams and 2. recovering target statistics by controlled numerical integration on a judiciously chosen grid. For $F_k$ with $k\in (0,1)$ we obtain the first one-pass streaming algorithms for Delphic set streams whose space and per-set update time are $\mathrm{poly}(\log|Ω|,\log m,\varepsilon^{-1},\log(1/δ))$ in the practically relevant regime $τ=\mathrm{polylog}(|Ω|,m)$; in general the bounds are polynomial in $τ$ and $\varepsilon^{-1}$ and logarithmic in $δ^{-1}$. We also give a complexity-theoretic barrier explaining why lower bounds for removing the bounded-frequency assumption appear difficult: ruling out polylogarithmic algorithms for unrestricted Delphic $F_k$ would imply a linear-space threshold-counting separation. △ Less

Submitted 18 August, 2026; v1 submitted 13 August, 2026; originally announced August 2026.

Comments: Fixed typos and a couple of localized proof fixes

Symbolic Machine Learning for Vapor-Liquid Equilibrium Prediction in Cx-N2 Binary Mixtures

Authors: Bongseok Kim, Suman Chakraborty, Gary Huang, Mehek Mathur, Guang Lin, Li Qiao

Abstract: Accurate prediction of vapor--liquid equilibrium (VLE) for hydrocarbon-nitrogen mixtures remains challenging for cubic equations of state, particularly across broad ranges of composition and hydrocarbon chain length. While deep learning models can provide accurate predictions, they often lack interpretability and explicit analytical expressions. In this work, we propose a symbolic machine learning… ▽ More Accurate prediction of vapor--liquid equilibrium (VLE) for hydrocarbon-nitrogen mixtures remains challenging for cubic equations of state, particularly across broad ranges of composition and hydrocarbon chain length. While deep learning models can provide accurate predictions, they often lack interpretability and explicit analytical expressions. In this work, we propose a symbolic machine learning approach to discover interpretable symbolic corrections to Peng-Robinson equation-of-state (PR-EOS) predictions from experimental data. The proposed approach adopts a two-level strategy: symbolic expressions are first identified for individual hydrocarbon systems, after which their coefficients are represented as functions of carbon number to enable accurate prediction across different hydrocarbon systems. The results demonstrate significantly improved prediction accuracy over the original PR-EOS across all hydrocarbon-nitrogen systems. Overall, the proposed approach provides an interpretable symbolic correction framework for improving PR-EOS predictions of hydrocarbon-nitrogen VLE. △ Less

Submitted 6 August, 2026; originally announced August 2026.

Toward Uncertainty Quantification in Modern Art

Authors: Tirtho Roy, Ushashi Bhattacharjee, Showrav Kumar Saha, Sayantan Chakraborty, Koushik Howlader, Tanusree Bhattacharjee

Abstract: Asked to animate the same modern artwork under different random seeds, a text to video model returns visibly different films, one reading per seed. Because modern art is ambiguous by intent, this disagreement is signal, not noise. Yet prevailing uncertainty quantification (UQ) collapses a set of generations to a dispersion scalar that says how much the seeds differ but not how: it cannot tell a co… ▽ More Asked to animate the same modern artwork under different random seeds, a text to video model returns visibly different films, one reading per seed. Because modern art is ambiguous by intent, this disagreement is signal, not noise. Yet prevailing uncertainty quantification (UQ) collapses a set of generations to a dispersion scalar that says how much the seeds differ but not how: it cannot tell a compact interpretation from a dominant reading plus an outlier, two competing modes, or diffuse instability, nor whether the set still contains a rendering faithful to the original. We present the first study of the structure of generative uncertainty for modern art animation, and a reusable protocol for identifying source blind multiseed uncertainty: a suite of seven source blind and six reference aware estimators; a distributional profile (robust spread, outlier influence, explicit topology, multimodality, anisotropy, leave one seed influence, reference coverage); a distribution model ablation (vMF, Kent, ACG, Student t, kernel, mixture); eight identification questions; and an artwork level statistical protocol. We build the first corpus: 250 modern artwork captions rendered by Wan2.1 14B under four seeds (1000 videos) across 4 encoders, artworks withheld from generation. As a diagnostic the protocol succeeds: it classifies seed set topology at balanced accuracy 0.98 (chance 0.25), isolates the outlier configuration at AUROC 1.00 where a scalar reaches only 0.35, and splits high uncertainty artworks into reference covering (n=97) and reference missing (n=56) diversity, reliably from three seeds and across encoders. △ Less

Submitted 3 August, 2026; originally announced August 2026.

HCCL: Collective Communication for Meta Training and Inference Accelerators

Authors: Wesley Bland, Tiago Antunes, Lars Paul Huse, Chidambaram Muthu, Adel Abouchaev, Rabib Alam, Abdullah Alperen, Alexey Andronov, Jose Anto Akkara, Vineet Badhwar, Pavan Balaji, Daniel Berkovitch, Bartosz Bogdanski, Shmeelok Chakraborty, Sungjun Cho, John Choi, James Custer, Rodrigo De Castro, Nguyen Dinh Pham, Matthew Edwards, Kristian Evensen, Evan Ezell, Alex Finestead, Seth Goldstein, Prankur Gupta , et al. (41 additional authors not shown)

Abstract: We present HCCL, a collective communication library co-designed with Meta's MTIA 300 accelerator, the first Meta chip to integrate backend networking directly on chip package. MTIA 300 includes dedicated message engines (MEs) with near-memory compute (NMC) that fully offload collective execution from the compute grid, enabling large overlap between computation and communication. HCCL uses a compil… ▽ More We present HCCL, a collective communication library co-designed with Meta's MTIA 300 accelerator, the first Meta chip to integrate backend networking directly on chip package. MTIA 300 includes dedicated message engines (MEs) with near-memory compute (NMC) that fully offload collective execution from the compute grid, enabling large overlap between computation and communication. HCCL uses a compiled communication model in which the host generates a complete description of each collective including dependencies. We describe the control and data path architecture, topology-aware algorithm selection across MTIA 300's asymmetric scale-up and scale-out network, and optimizations for both training and inference workloads. For training, HCCL achieves up to 940 GB/s on intra-rack collectives while introducing less than 0.5% degradation to concurrent compute throughput. For inference, we leverage one-sided communication primitives that bypass the scheduling path to minimize collective latency and describe collective designs that improve compute-communication pipelining for latency-sensitive workloads. △ Less

Submitted 31 July, 2026; originally announced August 2026.

Comments: 12 pages, 17 figures, to be published in the proceedings of "SC '26: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis"

The Case for Vibe Modeling: A Missing Step in AI-Based Trustworthy Software Development

Authors: Shalini Chakraborty, Michael Mittermaier, Judith Michael

Abstract: Large Language Models (LLMs) are increasingly used to generate software artifacts from natural language prompts. While this enables rapid prototyping and lowers the barrier to software creation, it also introduces challenges related to understanding, validation, traceability, and trust. In this paper, we argue that current AI-based development practices focus too heavily on the direct generation o… ▽ More Large Language Models (LLMs) are increasingly used to generate software artifacts from natural language prompts. While this enables rapid prototyping and lowers the barrier to software creation, it also introduces challenges related to understanding, validation, traceability, and trust. In this paper, we argue that current AI-based development practices focus too heavily on the direct generation of code and insufficiently on intermediate representations that preserve human intent and support reasoning about system behavior. We argue for vibe modeling as a lightweight intermediate abstraction between natural language interaction and code generation. To explore its potential, we present a student survey study that examines perceptions of LLM output understanding, validation effort, trust and the perceived usefulness of vibe modeling across several AI-assisted development scenarios. Our results are intended to inform future studies for trustworthy and explainable AI-based software engineering via vibe modeling. △ Less

Submitted 30 July, 2026; originally announced July 2026.

Comments: 7 pages, 1 figure

Estimating Size of the Union of Sets in Streaming Model

Authors: Kuldeep S. Meel, N. V. Vinodchandran, Sourav Chakraborty

Abstract: We study estimating the size of the union of sets $S_1,\dots,S_M$, where each $S_i\subseteqΩ$ is presented implicitly and arrives in a stream. We introduce Delphic sets, a class of streaming problems in which membership, sampling, and counting queries to each set are efficient, and show that this notion captures three well-known problems: Klee's measure problem (discrete version), test coverage es… ▽ More We study estimating the size of the union of sets $S_1,\dots,S_M$, where each $S_i\subseteqΩ$ is presented implicitly and arrives in a stream. We introduce Delphic sets, a class of streaming problems in which membership, sampling, and counting queries to each set are efficient, and show that this notion captures three well-known problems: Klee's measure problem (discrete version), test coverage estimation in combinatorial testing, and model counting of DNF formulas. Our primary contribution is a simple and efficient sampling-based algorithm that outputs an $(\varepsilon,δ)$-approximation of the cardinality of the union of Delphic sets in the streaming setting. It has space complexity $O(R\log|Ω|)$ and update time $O(R\log R\cdot\log(M/δ)\cdot\log|Ω|)$, where $R=O(\log(M/δ)\cdot\varepsilon^{-2})$. For the streaming Klee's measure problem, this gives the first algorithm whose update time depends linearly on the dimension $d$ for $d>1$, settling an open problem of Tirthapura and Woodruff (PODS 2012), and it directly yields efficient streaming algorithms for coverage estimation and DNF model counting. We further show that the space for coverage estimation can be made near-optimal at the cost of an update procedure in $\mathrm{P}^{\mathrm{NP}}$, revealing a time-space trade-off. A key strength of our approach is the simplicity of both the algorithm and its analysis, which makes it amenable to practical implementation. In this revised version, the algorithm and its correctness analysis have additionally been formalized and machine-checked in Lean 4. (Shortened for Arxiv) △ Less

Submitted 29 July, 2026; originally announced July 2026.

Comments: This is a significantly revised version of the paper that appeared in the proceedings of the 40th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (PODS-21). The main claims of the paper remain unchanged; however, we have fixed several typos and bugs in the proofs. The proofs and theorem statements have also been formalized in Lean

Dual-Level Atomic and Coordination Geometry Learning for Crystal Property Prediction Using Graph Neural Networks

Authors: Sanjay Chakraborty

Abstract: Accurate prediction of crystal properties remains a key challenge in computational materials science. While graph neural networks (GNNs) such as CGCNN, MEGNet, ALIGNN, and SchNet have shown strong performance, they primarily represent crystals at the atomic level and implicitly learn local chemical environments through message passing. However, many material properties are governed by coordination… ▽ More Accurate prediction of crystal properties remains a key challenge in computational materials science. While graph neural networks (GNNs) such as CGCNN, MEGNet, ALIGNN, and SchNet have shown strong performance, they primarily represent crystals at the atomic level and implicitly learn local chemical environments through message passing. However, many material properties are governed by coordination polyhedra, the fundamental structural units formed by atoms and their neighboring atoms. To address this limitation, we propose the Coordination Polyhedron Graph Network (CPGN), a multi-scale GNN that jointly learns atomic, bond, and coordination-polyhedron representations. CPGN constructs three coupled graphs: an atom graph encoding elemental and bonding information, a line graph capturing angular interactions, and a coordination polyhedron graph describing Voronoi-derived local environments through corner-, edge-, and face-sharing relationships. Physically meaningful geometric descriptors are incorporated for each polyhedron, while an interleaved message-passing mechanism with bidirectional cross-attention enables effective information exchange across structural levels. Extensive evaluations on the Materials Project, JARVIS-DFT, and QM9 benchmark datasets demonstrate that CPGN outperforms existing state-of-the-art GNN models. It achieves a formation-energy MAE of 0.060 eV/atom and a band-gap MAE of 0.292 eV on the Materials Project, while providing competitive multi-property prediction on JARVIS-DFT and superior HOMO prediction on QM9. The results highlight that explicit modeling of coordination polyhedra improves crystal representation learning and enables accurate, physically interpretable prediction of material properties. △ Less

Submitted 17 July, 2026; originally announced July 2026.

Motivations and Barriers to Communicating Software Engineering Research: Insights from Early Career Researchers

Authors: Shalini Chakraborty, Marvin Wyrich, Sven Apel, Sebastian Baltes

Abstract: Science communication is increasingly becoming a part of modern research careers, involving researchers to disseminate knowledge, engage broader communities, and increase the societal impact of their work. Despite its growing importance, little is known about how early-career software engineering researchers perceive and navigate science communication in practice. In this paper, we investigate how… ▽ More Science communication is increasingly becoming a part of modern research careers, involving researchers to disseminate knowledge, engage broader communities, and increase the societal impact of their work. Despite its growing importance, little is known about how early-career software engineering researchers perceive and navigate science communication in practice. In this paper, we investigate how PhD students in software engineering experience science communication. We conducted semi-structured interviews with 18 doctoral candidates from diverse international backgrounds. Using thematic analysis, we examine three interconnected dimensions: motivations, communication channels, and barriers. Our findings reveal a strong tension between aspiration and practice. Participants were highly motivated to engage in science communication due to opportunities for collaboration, professional recognition, broader impact, and advocacy for themselves and their research. However, translating these motivations into action was frequently constrained by social anxiety, uncertainty regarding appropriate audiences and communication venues, limited feedback mechanisms, insufficient institutional guidance, and challenges associated with navigating an increasingly fragmented communication landscape. Our findings highlight the need for tailored, balanced support systems that empower software engineering PhD students to engage in science communication effectively and confidently across diverse cultural and institutional environments. We outline practical implications that offer initial guidance for addressing these challenges in future work. △ Less

Submitted 27 July, 2026; originally announced July 2026.

Comments: 38 pages, 5 figures

Physics-Informed Neural Networks for Predicting Nitrous Oxide Flux

Authors: Freddy Yu, Jashanjeet Kaur Dhaliwal, Subhadeep Chakraborty

Abstract: Nitrous oxide (N$_2$O) is the dominant ozone-depleting substance emitted in the 21st century, and the third largest contributor to anthropogenic greenhouse gases due to its high potency and long atmospheric lifetime, with more than 70% of N$_2$O emissions occurring as a result of agricultural processes. Current approaches to predicting N$_2$O flux emissions include process-based models such as Day… ▽ More Nitrous oxide (N$_2$O) is the dominant ozone-depleting substance emitted in the 21st century, and the third largest contributor to anthropogenic greenhouse gases due to its high potency and long atmospheric lifetime, with more than 70% of N$_2$O emissions occurring as a result of agricultural processes. Current approaches to predicting N$_2$O flux emissions include process-based models such as DayCent and Cycles, as well as classical AI models, but the application of Physics-Informed Neural Networks (PINNs) to predicting N$_2$O flux emissions is largely underexplored. Our paper draws upon the mechanistic equations that underlie the DayCent family of process-based models to construct a rigorously derived, literature-traceable physics residual. We then build and train an MLP-based PINN on a multi-site agricultural dataset spanning four geographically distinct US agricultural sites. Across all tested values of the physics loss weighting hyperparameter $λ$, our PINN consistently and substantially outperformed uncalibrated Cycles simulation (R$^2=0.01$), with our MLP baseline achieving mean R$^2=0.411$ across ten random seeds. Physics constraints consistently degrade model performance in holdout validation, with marginal degradation at low $λ$ and significant degradation at high $λ$, but consistently improve model performance and reduce performance variability in leave-one-site-out validation. This suggests that physics constraints sacrifice in-distribution accuracy for out-of-distribution robustness, anchoring the model toward biogeochemically plausible behavior on unfamiliar soil conditions --- though cross-site generalization remains challenging, with negative R$^2$ across all seeds and $λ$ values on our geographically distinct held-out site. △ Less

Submitted 26 July, 2026; originally announced July 2026.

Comments: 28 pages, 7 figures

Pathologist Attention-Aligned Report Generation for Prostate Histopathology

Authors: Ruoyu Xue, Suryakant Singh, Souradeep Chakraborty, Pierre Marza, Oksana Yaskiv, Constantin Friedman, Natallia Sheuka, Paul Friedman, Bharat Ramlal, Beatrice Knudsen, Rajarsi Gupta, Joel Saltz, Prateek Prasanna, Gregory Zelinsky, Dimitris Samaras

Abstract: The allocation of visual attention by pathologists during cancer diagnosis is a highly selective process that critically shapes the information extracted from whole-slide images (WSIs). Human attention helps medical imaging tasks such as classification and segmentation, and becomes a strong semantic cue for identifying diagnostically informative regions for report generation. In this paper, we int… ▽ More The allocation of visual attention by pathologists during cancer diagnosis is a highly selective process that critically shapes the information extracted from whole-slide images (WSIs). Human attention helps medical imaging tasks such as classification and segmentation, and becomes a strong semantic cue for identifying diagnostically informative regions for report generation. In this paper, we introduce human attention into the training of pathologist report generation models. To this end, we collected a multimodal human-attention dataset of 121 prostate WSIs annotated with pathologists' multi-scale viewport trajectories synchronized with the pathologists' verbal descriptions and cursor movements for five clinically relevant components (e.g., Gleason patterns). Using this dataset, we finetune two report generation models with an attention-alignment loss that regularizes the model attention over image patches to match the distribution of pathologist attention. We evaluate our approach on prostate cancer report generation and visual question answering using two models with different internal attention mechanisms (i.e., how image tokens are integrated into the language decoder). Experiments show average gains of 10.9% on NLP-based metrics and 19.3% in accuracy across five clinically relevant report components. Further, model attention maps extracted at inference time, with minimal computational overhead, align more closely with pathologist attention, providing stronger visual support for the generated reports by highlighting the regions that most influence the output. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Comments: 11 pages, 4 figures, accepted for publication at the 29th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2026)

Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Cones

Authors: Ayoub Ghriss, Sourav Chakraborty

Abstract: Linear attention promises constant-time recurrent inference but degrades sharply on associative recall. We formulate attention recall as a spherical-packing problem and introduce Kernelized Linear Attention Activations (KATA), a framework whose feature maps are derived from first principles by certifying nonnegative attention weights through a self-dual homogeneous cone. Building on this observati… ▽ More Linear attention promises constant-time recurrent inference but degrades sharply on associative recall. We formulate attention recall as a spherical-packing problem and introduce Kernelized Linear Attention Activations (KATA), a framework whose feature maps are derived from first principles by certifying nonnegative attention weights through a self-dual homogeneous cone. Building on this observation, we show that rank-one positive semi-definite (PSD) features offer a favorable capacity--interference tradeoff. KATA recovers a parameter-free convex output gate and characterizes associative capacity through the Welch interference floor. For tolerances above this floor, KATA enlarges the state without adding parameters and admits spherical codes with exponentially many keys in the projection dimension. We implement KATA as fused Triton kernels at two operating points: a flash-attention-style forward up to ${\sim}1.6\times$ FlashAttention-2 throughput, and an exact $O(T)$ chunked-state form that reaches ${\sim}11\times$ FlashAttention-2 forward throughput at $131$k tokens. An associative scan of the first-order feature lowers the inter-chunk recurrence depth to $O(\log(T/C))$ for chunk size $C$ and averages ${\sim}2.4\times$ the throughput of a matched sequential linear-attention baseline. On long-range MQAR and repeated-key overwrite, several KATA variants outperform Gated DeltaNet, with parameter counts and state sizes reported alongside accuracy. Induction preserves near-perfect recall, while kernel benchmarks show that the maps can be implemented efficiently. KATA retains $0.985$ MQAR at a $16\times$ out-of-distribution length, approaching the softmax with roughly one quarter of the KV-cache entries. Experiments on 340M-parameter LLMs reveal a feature-dependent fluency trade-off and clarify how positional embeddings, delta rules, and decay gates interact with feature geometry. △ Less

Submitted 19 July, 2026; originally announced July 2026.

Comments: 39 pages, 4 figures, 10 tables. Code: https://github.com/ayghri/kata

MSC Class: 68T07; 52C17 ACM Class: I.2.6; G.3

A Quantum-Classical Hybrid Framework for Multivariate Time-Series Forecasting Complexity-Fidelity Trade-offs and Limitations

Authors: Sanjay Chakraborty, Fredrik Heintz

Abstract: This paper presents a unified quantum-classical hybrid framework for multi-horizon time-series forecasting, introducing two model variants Quantum Reservoir Forecaster (QRC-F) and Variational Quantum Forecaster (VQF-F). The proposed framework investigates the complexity-fidelity trade-off of quantum forecasting under near-term NISQ hardware constraints. Continuous time-series signals are transform… ▽ More This paper presents a unified quantum-classical hybrid framework for multi-horizon time-series forecasting, introducing two model variants Quantum Reservoir Forecaster (QRC-F) and Variational Quantum Forecaster (VQF-F). The proposed framework investigates the complexity-fidelity trade-off of quantum forecasting under near-term NISQ hardware constraints. Continuous time-series signals are transformed into binary representations through uniform quantization and encoded into quantum states using angle encoding with parameterized RY rotation gates. Cross-channel entanglement layers capture dependencies among multiple variables. QRC-F utilizes a fixed random unitary quantum reservoir for stable, gradient-free temporal feature extraction, whereas VQF-F employs a trainable variational quantum circuit optimized through the parameter-shift rule to learn temporal and inter-variable patterns from Pauli expectation values. Both models replace computationally expensive quadratic self-attention with efficient linear transformations, reducing parameter complexity. A shared MIMO-based multi-horizon prediction head simultaneously generates forecasts across multiple horizons, avoiding error accumulation in recursive forecasting. Experimental evaluations on benchmark datasets, including ETTh1, ETTh2, ETTm1, ETTm2, Weather, electricity, and exchange-rate, demonstrate that VQF-F achieves superior training stability and parameter efficiency, while QRC-F provides enhanced robustness and circuit fidelity under quantum noise. The results establish a practical quantum-native forecasting framework with strong potential for deployment on near-term NISQ devices. △ Less

Submitted 17 July, 2026; originally announced July 2026.

Learning from Local Walks on Dynamic Graphs with Bandit Feedback

Authors: Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni, Lijun Chen

Abstract: We study stochastic multi-armed bandits on dynamic graphs, where arms correspond to the vertices of a network with time-varying edges. In this setting, the learner is restricted to local movement, selecting only its current node or an immediate neighbor at each round. This constraint decouples best-arm identification from exploitation: even after the optimal arm is identified, the learner may rema… ▽ More We study stochastic multi-armed bandits on dynamic graphs, where arms correspond to the vertices of a network with time-varying edges. In this setting, the learner is restricted to local movement, selecting only its current node or an immediate neighbor at each round. This constraint decouples best-arm identification from exploitation: even after the optimal arm is identified, the learner may remain unable to reach it through the evolving topology. We identify a process-agnostic structural condition, based on sliding-window mixing, that ensures the graph's intrinsic walk remains stable for both exploration and navigation. Under this regime, we analyze a family of local explore-then-commit algorithms and establish sublinear expected regret. Our framework includes a reward-aware strategy, for which we prove a worst-case safety theorem and a separate performance gain theorem. △ Less

Submitted 27 July, 2026; v1 submitted 12 July, 2026; originally announced July 2026.

Trusting AI to increase productivity? Perspectives Across the Global North and South

Authors: Adam Bokun, Shalini Chakraborty

Abstract: Generative AI (GenAI) tools are widely used in academia and software development, where productivity gains may depend not only on technical capabilities but also on users' trust and contextual factors. This paper presents emerging results from an exploratory study investigating the relationship between trust in GenAI and perceived productivity, motivated by Global South contexts. We conducted a sy… ▽ More Generative AI (GenAI) tools are widely used in academia and software development, where productivity gains may depend not only on technical capabilities but also on users' trust and contextual factors. This paper presents emerging results from an exploratory study investigating the relationship between trust in GenAI and perceived productivity, motivated by Global South contexts. We conducted a systematic literature review, complemented by a grey literature analysis and a survey study. The literature review identified no peer-reviewed evidence at the intersection of GenAI trust, productivity, and Global South settings, while the grey literature revealed only limited insights. At the time of writing, the survey has received 36 valid responses from participants across both the Global North and Global South, including individuals with cross-regional experiences. Preliminary results suggest that respondents born and working in the Global South tended to trust AI more, but did not usually report clear productivity gains from using it. In contrast, respondents born and working outside the Global South reported stronger productivity gains and greater time savings, even though they showed less trust in generative AI. These findings suggest that trusting AI is not enough on its own; productivity also depends on access, the type of task, and how much users need to check the output. △ Less

Submitted 11 July, 2026; originally announced July 2026.

Structured Thoughts For Improved Reasoning And Context Pruning

Authors: Zain Sarwar, Supriyo Chakraborty, Berkcan Kapusuzoglu, Chia-Hsuan Lee, Anirban Das, Stephen Rawls, Kartik Balasubramaniam, Sambit Sahu

Abstract: Large language models (LLMs) excel at generating long chains of thought, but long reasoning traces are often verbose and memory-inefficient. In this work, we introduce Structured Thoughts, a framework that organizes reasoning into alternating <try> and <outcome> blocks: <try> captures exploratory scratch work, while <outcome> contains the distilled conclusion of that step. We construct a dataset o… ▽ More Large language models (LLMs) excel at generating long chains of thought, but long reasoning traces are often verbose and memory-inefficient. In this work, we introduce Structured Thoughts, a framework that organizes reasoning into alternating <try> and <outcome> blocks: <try> captures exploratory scratch work, while <outcome> contains the distilled conclusion of that step. We construct a dataset of structured thoughts by segmenting reasoning traces into <try> blocks and prompting an LLM to summarize each step into its corresponding <outcome>. Fine-tuning pretrained foundation models on this reformatted data produces models that adopt the structured reasoning style, leading to performance gains of up to 8.08\% on reasoning benchmarks compared to standard SFT. The explicit structure also enables context pruning: after each <try>/<outcome> pair, the <try> can be pruned, allowing the model to retain conclusions without keeping the full scratch work in the context. A proof-of-concept pruning implementation achieves an average of 85\% memory / context savings with an 8.67\% performance drop across mathematical tasks. △ Less

Submitted 11 July, 2026; originally announced July 2026.

Computing over Data Streams using Catalytic Space

Authors: Ripley Becker, Sourav Chakraborty, Debarshi Chanda, A. Pavan, N. V. Vinodchandran

Abstract: We introduce a streaming model with \emph{catalytic memory}, an auxiliary workspace that must be returned to its initial state at the end of the computation. We show that catalytic space yields dramatic space savings for data stream algorithms. We first study the exact computation of frequency moments in insertion-only data streams. For every $k\ge1$, we give an exact four-pass algorithm for compu… ▽ More We introduce a streaming model with \emph{catalytic memory}, an auxiliary workspace that must be returned to its initial state at the end of the computation. We show that catalytic space yields dramatic space savings for data stream algorithms. We first study the exact computation of frequency moments in insertion-only data streams. For every $k\ge1$, we give an exact four-pass algorithm for computing $\mathbb{F}_{k}$ using $O(k\log m)$ clean space, where $m$ is the stream length. We also present a $(k+1)$-pass algorithm with the same clean-space complexity that uses a factor of $k$ less catalytic space than the four-pass algorithm. For small moments, we obtain stronger results. In particular, we show that $\mathbb{F}_{2}$ and $\mathbb{F}_{3}$ can be computed exactly in two and three passes, respectively, using only $O(\log m)$ clean space. Additionally, we show that exact $\mathbb{F}_{0}$ computation reduces to computing $\mathbb{F}_{k}$ for a suitably chosen large value of $k$, resulting in an exact four-pass algorithm for $\mathbb{F}_{0}$ using only $O(\log m)$ clean space. We further show how our frequency-moment algorithms can be used to exactly count induced occurrences of any fixed graph $H$ in a graph stream, yielding a four-pass algorithm that uses $O_H(\log n)$ clean space, where $n$ is the number of vertices in the graph. As a special case, we obtain an exact three-pass algorithm for triangle counting using $O(\log n)$ clean space. All of our algorithms are multi-pass. We complement these algorithmic results with a matching limitation showing that catalytic memory does not provide additional power in the single-pass setting. Specifically, we prove that every randomized or deterministic single-pass streaming algorithm using $s$ bits of clean memory and catalytic space can be simulated in the standard streaming model, without catalytic memory, using $O(s)$ space. △ Less

Submitted 9 July, 2026; originally announced July 2026.

CaSPECT: Discovering Causally Homogeneous Subgroups via Directed Spectral Clustering

Authors: Arghya Pratihar, Shinjon Chakraborty, Swagatam Das

Abstract: We propose \textbf{CaSPECT}, a causal spectral clustering framework for discovering causally homogeneous subgroups from observational data. Rather than clustering in covariate space, CaSPECT defines similarity through the topology of a learned directed acyclic graph (DAG); a bootstrap-stabilised PC algorithm recovers the causal skeleton; a novel \emph{Orientation Validation Score} (OVS) combines P… ▽ More We propose \textbf{CaSPECT}, a causal spectral clustering framework for discovering causally homogeneous subgroups from observational data. Rather than clustering in covariate space, CaSPECT defines similarity through the topology of a learned directed acyclic graph (DAG); a bootstrap-stabilised PC algorithm recovers the causal skeleton; a novel \emph{Orientation Validation Score} (OVS) combines PC bootstrap evidence with DirectLiNGAM to orient edges robustly; directed edges are weighted by backdoor-identified average treatment effects estimated via OLS or double machine learning. Chung's directed Laplacian provides a spectral embedding in which individuals close together share the same causal propagation pathways. We establish almost-sure consistency of the full pipeline and validate the method through a controlled simulation study and on LaLonde CPS1, IHDP, and 401(k) datasets, where CaSPECT recovers a positive and statistically significant treatment effect within the causally comparable subpopulation and corrects for severe confounding without requiring a pre-specified propensity score model. △ Less

Submitted 3 July, 2026; originally announced July 2026.

Diagnosing Aerial-View Object Detectors with Foundational Image Generative Models

Authors: Stanislav Panev, Minhyek Jeon, Vaishnavi Khindkar, Ahish Deshpande, Celso M de Melo, Shuowen Hu, Shayok Chakraborty, Fernando De la Torre

Abstract: Recent advances in large-scale image generative models enable photorealistic scene synthesis with controllable attributes. Beyond data augmentation, their potential as diagnostic tools for trained vision systems remains unexplored in the aerial and remote sensing domains. We introduce a synthetic diagnostic framework for aerial-view vehicle detection that combines text-guided generation, attribute… ▽ More Recent advances in large-scale image generative models enable photorealistic scene synthesis with controllable attributes. Beyond data augmentation, their potential as diagnostic tools for trained vision systems remains unexplored in the aerial and remote sensing domains. We introduce a synthetic diagnostic framework for aerial-view vehicle detection that combines text-guided generation, attribute-controlled editing, and automated attribute verification to construct a controllable synthetic testbed. This enables fine-grained evaluation of pretrained detectors under diverse scene types and environmental conditions that are difficult to isolate in real datasets. Across three detection architectures and three real aerial datasets, synthetic scene-wise performance trends closely match real-world weaknesses. Guided by these diagnostics, targeted supplementation with small real datasets from the identified weak categories yields improvements of up to 13% AP50 while requiring substantially fewer additional samples than non-targeted augmentation. Our results show that controlled synthetic probing can predict real-domain performance gaps and guide efficient data collection. The proposed diagnostic framework is modular and can incorporate alternative generative or vision-language models as capabilities evolve. Our code and datasets are available here: https://humansensinglab.github.io/AVODDiag/ △ Less

Submitted 2 July, 2026; originally announced July 2026.

On the Policy Gradient Foundations of Group Relative Policy Optimization: Credit Assignment, Gradient Sparsity, and Rank Collapse

Authors: Amritansh Mishra, Supriyo Chakraborty, Berkcan Kapusuzoglu

Abstract: Group Relative Policy Optimization (GRPO) eliminates the learned critic in PPO by using the mean reward of grouped rollouts as a baseline. We provide a rigorous derivation of GRPO from first principles of the policy gradient theorem, revealing a fundamental credit assignment failure: under output-only reward, every token in a rollout receives identical advantage, collapsing token-level credit to a… ▽ More Group Relative Policy Optimization (GRPO) eliminates the learned critic in PPO by using the mean reward of grouped rollouts as a baseline. We provide a rigorous derivation of GRPO from first principles of the policy gradient theorem, revealing a fundamental credit assignment failure: under output-only reward, every token in a rollout receives identical advantage, collapsing token-level credit to a single scalar. We prove this induces gradient sparsity that intensifies over training, and demonstrate empirically via SVD analysis of GRPO gradients on Nemotron-4B/GSM8K that the gradient matrix has effective rank $\approx$ 2 regardless of group size $R \in \{2, 4, 8\}$. We formalize this as an intrinsic rank-2 structure arising from the zero-sum constraint on advantages and derive conditions under which GRPO's baseline is optimal. Our results characterize when GRPO's simplicity is theoretically justified and identify the credit assignment bottleneck as the key limitation for multi-step reasoning. △ Less

Submitted 28 June, 2026; originally announced June 2026.

A Physics-Informed Fourier-Wavelet Transformer for Multiscale Computational Fluid Dynamics Surrogate Modeling

Authors: Somyajit Chakraborty, Ming Pan, Xizhong Chen

Abstract: Physics-informed surrogate models can accelerate computational fluid dynamics simulations. However, many existing methods reproduce global flow patterns more reliably than localized multiscale structures. This study presents a physics-informed Fourier-wavelet transformer for next-step velocity-field reconstruction in real-world flow benchmarks. The proposed formulation combines hybrid Fourier-wave… ▽ More Physics-informed surrogate models can accelerate computational fluid dynamics simulations. However, many existing methods reproduce global flow patterns more reliably than localized multiscale structures. This study presents a physics-informed Fourier-wavelet transformer for next-step velocity-field reconstruction in real-world flow benchmarks. The proposed formulation combines hybrid Fourier-wavelet spectral encoding with physics-biased self-attention based on partial differential equation residual diagnostics. It also uses self-supervised pretraining through Masked Physics Prediction and Equation Consistency Prediction. The experiments are conducted on two real benchmark cases: cylinder-wake flow and fluid-structure interaction. All approaches are evaluated under a shared local protocol and compared with spectral, transformer-based, operator-learning, and physics-informed neural-network baselines. On the cylinder-wake benchmark, the proposed model achieves the best aggregate accuracy, with an all-channel normalized mean-squared error of 0.05875 and an all-channel Pearson correlation coefficient of 0.97019. On the fluid-structure-interaction benchmark, it gives the lowest all-channel normalized mean-squared error of $2.70 \times 10^{-4}$, compared with $4.02 \times 10^{-4}$ for the strongest baseline. Component-wise field comparisons and scale-separated diagnostics further show stronger recovery of localized wake structures, including near-body, wake-core, and far-wake features. The results demonstrate improved real-world flow reconstruction while maintaining a practical accuracy-cost tradeoff. △ Less

Submitted 23 June, 2026; originally announced June 2026.

ACM Class: I.2.6; I.6.5; J.2

Towards Data-Efficient Cross-Device Generalization of Grad-Shafranov Equilibria via Transfer Learning Neural Operator

Authors: Jay Phil Yoo, William Howes, Yashika Ghai, Kazuma Kobayashi, Souvik Chakraborty, Syed Bahauddin Alam

Abstract: Real-time reconstruction of magnetohydrodynamic equilibria is essential for plasma shaping, stability assessment and feedback control in magnetic confinement fusion. However, Grad-Shafranov equilibrium calculations remain largely device-specific and iterative, limiting their use in latency-constrained control settings. Existing neural approaches can accelerate individual equilibrium predictions, b… ▽ More Real-time reconstruction of magnetohydrodynamic equilibria is essential for plasma shaping, stability assessment and feedback control in magnetic confinement fusion. However, Grad-Shafranov equilibrium calculations remain largely device-specific and iterative, limiting their use in latency-constrained control settings. Existing neural approaches can accelerate individual equilibrium predictions, but they do not generally provide reusable models across changing plasma boundaries or tokamak geometries. Here we show that equilibrium reconstruction can be recast as a cross-device operator learning problem. We develop a domain-specific neural operator framework that maps geometry and profile parameters directly to the poloidal flux field, replacing repeated solve-on-demand computation with amortized operator inference. Using the analytically tractable Solov'ev family as a controlled Grad-Shafranov testbed, we generate equilibria across eight geometrically distinct tokamak-like configurations and benchmark five neural operator architectures under four transfer-learning strategies. Single-geometry pretraining gives poor transfer to unseen devices, whereas multi-geometry pretraining enables data-efficient adaptation. The Wavelet Neural Operator gives the strongest cross-geometry performance, reaching mean relative L2 errors below 4% with 100 labelled target equilibria and below 2% with full fine-tuning. The predicted magnetic fields satisfy the divergence-free constraint to numerical precision, and four architectures achieve millisecond or sub-millisecond inference. These results identify neural operator pretraining as a route towards reusable, real-time equilibrium inference across fusion device configurations. △ Less

Submitted 13 June, 2026; originally announced June 2026.

Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities

Authors: Dipto Das, Achhiya Sultana, Ankit Singh Chauhan, Saadia Binte Alam, Mohammad Shidujaman, Shion Guha, Sunandan Chakraborty, Syed Ishtiaque Ahmed

Abstract: Language operates as a mechanism of both marginalization and resistance, especially for minority communities navigating insensitive and harmful speech online. As content moderation increasingly depends on large language models (LLMs), concerns arise about whether these systems can recognize culturally insensitive speech-language that disregards or marginalizes the cultural and religious perspectiv… ▽ More Language operates as a mechanism of both marginalization and resistance, especially for minority communities navigating insensitive and harmful speech online. As content moderation increasingly depends on large language models (LLMs), concerns arise about whether these systems can recognize culturally insensitive speech-language that disregards or marginalizes the cultural and religious perspectives of historically underrepresented communities, often through implicit erasure, misrepresentation, or normative framing, rather than overt hostility. Focusing on Bangladesh's Hindu and Chakma communities -- the country's largest religious and Indigenous ethnic minorities, respectively -- this paper investigates the epistemic limits of LLM-based moderation systems and explores methods for incorporating minority perspectives. We co-created a culturally grounded corpus of insensitive speech with community members and integrated their narratives into moderation pipelines using retrieval augmented generation (RAG). Our tool, Mod-Guide, improves LLM sensitivity to minority viewpoints by leveraging contextual cues derived from lived experience. Through mixed-method evaluations involving both minority and majority participants, we demonstrate that RAG-enhanced moderation responses are more contextually accurate and perceived differently across ethnic lines. This work advances research in human-computer interaction, AI ethics, and social computing by foregrounding restorative justice and hermeneutical inclusion in the design of content moderation systems. △ Less

Submitted 11 June, 2026; originally announced June 2026.

FlowBank: Query-Adaptive Agentic Workflows Optimization through Precompute-and-Reuse

Authors: Lingzhi Yuan, Chenghao Deng, Fangxu Yu, Souradip Chakraborty, Mohammad Rostami, Furong Huang

Abstract: Large Language Model (LLM)-based multi-agent systems are increasingly powerful, but current agentic workflow optimization paradigms make an unsatisfying trade-off. Task-level methods spend substantial offline compute yet deploy only a single workflow, leaving complementary candidates unused, while query-level methods synthesize a new workflow per query at substantial inference cost. Our motivating… ▽ More Large Language Model (LLM)-based multi-agent systems are increasingly powerful, but current agentic workflow optimization paradigms make an unsatisfying trade-off. Task-level methods spend substantial offline compute yet deploy only a single workflow, leaving complementary candidates unused, while query-level methods synthesize a new workflow per query at substantial inference cost. Our motivating analysis shows these paradigms are more complementary than competing: workflows discovered during offline search often solve different subsets of queries, and many queries handled by expensive query-level generation can already be solved by cheaper precomputed workflows. This suggests a different objective: rather than searching for one universally best workflow or regenerating one per instance, we should build a compact bank of reusable, complementary workflows and select among them adaptively at inference time. Doing so requires solving three coupled problems: generating complementary rather than redundant candidates, compressing them into a small deployable portfolio, and assigning each query to the right workflow under a performance-cost trade-off. To this end, we present FlowBank, a three-stage framework for portfolio-based agentic workflow optimization. Diversifying proposes DiverseFlow to steer search toward under-covered queries and produce a high-coverage candidate pool. Curating proposes CuraFlow to compress this pool into a compact portfolio with minimal redundancy. Matching casts deployment as edge-value prediction on a query-workflow bipartite graph and routes each incoming query to the portfolio member with the best predicted utility. Across five benchmarks, FlowBank achieves the highest average score among the evaluated methods while remaining cost-competitive, improving over the strongest automated and handcrafted baselines by 4.26% and 14.92% relative, respectively. △ Less

Submitted 9 June, 2026; originally announced June 2026.

A History-Aware Visually Grounded Critic for Computer Use Agents

Authors: Jaewoo Lee, Zaid Khan, Archiki Prasad, Justin Chih-Yao Chen, Supriyo Chakraborty, Kartik Balasubramaniam, Sambit Sahu, Elias Stengel-Eskin, Hyunji Lee, Mohit Bansal

Abstract: Various test-time interventions for Computer Use Agents (CUAs), including critic models, have been developed to improve performance through pre-execution action evaluation in complex Graphical User Interface (GUI) environments. However, existing critics suffer from two key limitations: they (1) focus primarily on short-sighted decision loops (e.g., forgetting earlier actions) and (2) lack the visu… ▽ More Various test-time interventions for Computer Use Agents (CUAs), including critic models, have been developed to improve performance through pre-execution action evaluation in complex Graphical User Interface (GUI) environments. However, existing critics suffer from two key limitations: they (1) focus primarily on short-sighted decision loops (e.g., forgetting earlier actions) and (2) lack the visual grounding needed to detect flawed actions (e.g., clicking wrong UI elements). To address these, we introduce HiViG, a History-aware Visually Grounded test-time framework, built around a multimodal critic trained on real GUI trajectories to abstract past interactions into a compact record and to evaluate actions with visual grounding. At test time, HiViG integrates the critic into the policy decision loop to provide macro-action history, which summarizes the policy's completed achievements, and visually grounded critique, which verifies raw execution coordinates against the current screenshot to intercept errors before execution. Across web, mobile, and desktop benchmarks, HiViG consistently outperforms existing scalar and verbal critics, improving average success rates over the strongest baseline by 5.8% for Qwen3-VL-32B and 9.0% for Gemini-3-Flash, and demonstrates strong cross-platform generalization. Ablations show that macro-action history mitigates short-sighted planning and visually grounded critique reduces execution errors, with both components being critical for test-time scaling in long-horizon GUI tasks. △ Less

Submitted 9 June, 2026; originally announced June 2026.

Comments: Code: https://github.com/G-JWLee/HiViG

Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models

Authors: Prajakta Kini, Avinash Reddy, Souradip Chakraborty, Satya Sai Srinath Namburi GNVV, Furong Huang, Amrit Singh Bedi, Alvaro Velasquez

Abstract: Instruction-tuned LLMs are increasingly converted into reasoning models through post-training to improve multi-step task performance. This conversion is usually optimized for reasoning accuracy, without explicitly preserving the alignment behavior of the instruction-tuned model, such as safe refusal, bias avoidance, and privacy protection. We ask: does this conversion preserve alignment? We study… ▽ More Instruction-tuned LLMs are increasingly converted into reasoning models through post-training to improve multi-step task performance. This conversion is usually optimized for reasoning accuracy, without explicitly preserving the alignment behavior of the instruction-tuned model, such as safe refusal, bias avoidance, and privacy protection. We ask: does this conversion preserve alignment? We study this question through a trustworthiness audit and find that it is not behavior-preserving by default. For a systematic analysis, we compare reasoning models produced via supervised fine-tuning, RL-based post-training, and distillation against matched instruction-tuned baselines across six trustworthiness dimensions: safety, toxicity, stereotyping and bias, machine ethics, privacy, and out-of-distribution robustness. We observe that reasoning models often improve on reasoning benchmarks but exhibit alignment regressions, including increased toxicity, amplified stereotyping, miscalibrated refusal, and contextual privacy leakage. These regressions are consistent with behavioral drift from the instruction-tuned baseline, measured by KL divergence. Overall, our results point to the broader conclusion that trustworthiness metrics are essential for evaluating reasoning models and should be reported alongside gains in reasoning capability. △ Less

Submitted 9 June, 2026; originally announced June 2026.

TRAPS: Treatment-Assignment Prediction via Pathway-informed Stratification

Authors: Sujoy Banik, Sayantan Chakraborty, Boishakhi Das Toma, Zainab Ghafoor, Ushashi Bhattacharjee, Koushik Howlader, Tirtho Roy

Abstract: Cancer treatment involves decisions across multiple clinical outcomes, yet pathway-informed deep learning models are typically evaluated in isolation, making their relative benefits unclear. We present a harmonized benchmark of three biologically informed architectures, BINN, GraphPath, and PATH, for predicting treatment exposure and short-term survival across five TCGA cancer cohorts comprising 2… ▽ More Cancer treatment involves decisions across multiple clinical outcomes, yet pathway-informed deep learning models are typically evaluated in isolation, making their relative benefits unclear. We present a harmonized benchmark of three biologically informed architectures, BINN, GraphPath, and PATH, for predicting treatment exposure and short-term survival across five TCGA cancer cohorts comprising 2,622 patients represented by Reactome pathway activity scores. Treatment labels indicate recorded exposure in TCGA rather than therapeutic response. All models jointly predict targeted molecular therapy (TMT), radiation therapy (RT), and six-month overall survival (OS) from a shared pathway representation and are evaluated on identical stratified folds using five repeated splits and paired-bootstrap testing. Under this controlled evaluation, most differences between architectures fall within 95 percent confidence intervals, indicating that rankings suggested by isolated evaluations are largely not statistically resolved. The main exception is survival prediction: the sparse-hierarchy BINN significantly outperforms both graph models on breast-cancer OS, with an AUROC improvement of up to 0.14 and p less than or equal to 0.01, and leads on lung and prostate OS. For treatment exposure, TMT is best discriminated in prostate cancer, with AUROC approximately 0.80 for all models, but no architecture significantly outperforms another on any TMT cohort. RT prediction remains weak across models, suggesting that its determinants may be more clinical than transcriptomic. Overall, architecture choice has limited impact under a unified evaluation, while short-term survival provides the clearest differentiation among pathway-informed models. △ Less

Submitted 10 August, 2026; v1 submitted 5 June, 2026; originally announced June 2026.

Report number: 4th International Conference on Computing Advancements (ICCA) 2026

The Hidden Bias of Process Reward Models:PRISM for Rewarding the Right Reasoning

Authors: Aakriti Agrawal, Souradip Chakraborty, Armin Saghafian, Nihal Sharma, Rizal Fathony, Nam H Nguyen, C. Bayan Bruss, Amrit Singh Bedi, Furong Huang

Abstract: Process Reward Models (PRMs) improve credit assignment for reasoning by providing step-level feedback. However, we identify a hidden bias in PRMs caused by severe imbalance in step-level training data. Standard cross-entropy training amplifies this bias, causing PRMs to overcredit plausible but incorrect steps and produce high false-positive rates. We show that these false positives have an asymme… ▽ More Process Reward Models (PRMs) improve credit assignment for reasoning by providing step-level feedback. However, we identify a hidden bias in PRMs caused by severe imbalance in step-level training data. Standard cross-entropy training amplifies this bias, causing PRMs to overcredit plausible but incorrect steps and produce high false-positive rates. We show that these false positives have an asymmetric downstream effect: false negatives mainly slow exploration, whereas false positives actively steer Best-of-N selection, guided decoding, and policy optimization toward flawed reasoning. This suggests that PRM training should shift from pointwise label fitting to reliable relative comparisons. To address this, we propose PRISM (Precision Ranking for Improved Step Modeling), a policy-aware PRM training framework that learns from contrastive step-level comparisons and hard negatives generated by a temporal lookahead strategy, requiring no new human labels. We further use a difficulty-aware curriculum to optimize the contrastive step margin. Across PRMBench and ProcessBench, PRISM substantially reduces false positives (22% on PRMBench) and improves macro F1 over strong discriminative PRMs. When applied to policy optimization and search tasks, including guided decoding and Best-of-N selection, it consistently improves accuracy (up to 22% for guided decoding and 33% for Best-of-N) and robustness. More broadly, trustworthy process supervision is not just about assigning high rewards, but about rewarding the right reasoning for the right reasons. △ Less

Submitted 8 June, 2026; originally announced June 2026.

DAL-PCQA: Enabling Distortion-Level and Language-Driven Reasoning for Point Cloud Quality Assessment

Authors: Swarna Chakraborty, Gabriel De Castro Araújo, Syeda Tasmi Faria, Marcelo M. Carvalho, Mylene C. Q. Farias

Abstract: Point Cloud Quality Assessment (PCQA) methods typically predict scalar Mean Opinion Scores (MOS), which quantify overall perceptual degradation but do not reveal its causes. In contrast, human observers naturally reason in terms of specific distortions such as blur, color shifts, point density changes, missing regions, and geometric deformations. To close this gap, we introduce DAL-PCQA, a distort… ▽ More Point Cloud Quality Assessment (PCQA) methods typically predict scalar Mean Opinion Scores (MOS), which quantify overall perceptual degradation but do not reveal its causes. In contrast, human observers naturally reason in terms of specific distortions such as blur, color shifts, point density changes, missing regions, and geometric deformations. To close this gap, we introduce DAL-PCQA, a distortion-aware, language-annotated dataset for PCQA. DAL-PCQA augments benchmark point clouds with multi-level distortion severity labels, discrete quality categories, and structured natural language descriptions aligned with human perception. We define a point-cloud-specific distortion taxonomy that covers both photometric and geometric artifacts. Statistical analysis reveals characteristic degradation patterns across distortion types and quality levels. To assess the utility of these annotations, we compare zero-shot and fine-tuned multimodal models for generating perceptual quality descriptions. Experiments show that distortion-aware supervision substantially improves lexical and semantic alignment with ground-truth descriptions. By enabling interpretable, distortion-level reasoning, DAL-PCQA facilitates language-driven, explainable point cloud quality assessment. The dataset is publicly available at https://github.com/swarna96/DAL-PCQA. △ Less

Submitted 5 June, 2026; originally announced June 2026.

Comments: Accepted at Qomex 2026

Domain Diversity, Motivation, Inclusion, and Feedback in Software Modelling Education

Authors: Isabella Graßl, Christopher Lazik, Shalini Chakraborty, Grischa Liebel, Miguel Goulão

Abstract: Student engagement is critical for effective learning in software modelling, yet fostering motivation and inclusivity remains a challenge. While existing research has focused on modelling tools, notations, and assessment, little attention has been given to how the choice of problem domains and the diversity, relatability, and cultural perspectives they bring shape students' learning experiences. T… ▽ More Student engagement is critical for effective learning in software modelling, yet fostering motivation and inclusivity remains a challenge. While existing research has focused on modelling tools, notations, and assessment, little attention has been given to how the choice of problem domains and the diversity, relatability, and cultural perspectives they bring shape students' learning experiences. This study explores how problem domains and teaching methods influence motivation, engagement, inclusiveness, and feedback in modelling education. To investigate these dimensions, we conducted parallel surveys with 90 students and 22 educators. Our findings reveal disconnects between educator assumptions and student preferences: Students show greatest motivation for socially relevant domains and prefer choice in selection, while educators overestimate interest in study-related domains. The study identifies how minor design choices can exclude students. Students perceive feedback as meaningful when visibly acted upon. These findings suggest inclusive domain selection is central to student motivation; thus, we recommend student-centred domain selection. △ Less

Submitted 4 June, 2026; originally announced June 2026.

Comments: Preprint. To be published in the 34th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (FSE Companion '26), July 05-09, 2026, Montreal, QC, Canada. ACM, New York, NY, USA, 12 pages

MSC Class: D.2.2; K.3.2

A differentiable machine learning small-angle X-ray scattering analysis framework for structure elucidation of lipid nanoparticles

Authors: Maria Bånkestad, Sandra Barman, Magnus Röding, Erik Kaunisto, Viktoriia Meklesh, Audrey Gallud, Marco Mendez, Marianna Yanez Arteta, Stefan Norberg, Ann Terry, Smita Chakraborty, Shun Yu, Jerk Rönnols, Sepideh Pashami

Abstract: Lipid nanoparticles (LNPs) are efficient delivery systems for negatively charged nucleic acids. Their multi-component architecture yields a core-shell structure. Small-angle X-ray scattering (SAXS) is an important characterization technique for LNPs, but recovering internal structure and size distribution from SAXS is an inverse problem with non-unique solutions. Realistic models are often too exp… ▽ More Lipid nanoparticles (LNPs) are efficient delivery systems for negatively charged nucleic acids. Their multi-component architecture yields a core-shell structure. Small-angle X-ray scattering (SAXS) is an important characterization technique for LNPs, but recovering internal structure and size distribution from SAXS is an inverse problem with non-unique solutions. Realistic models are often too expensive for systematic exploration. We introduce a machine-learning-accelerated, differentiable framework for SAXS analysis of heterogeneous, polydisperse LNPs. The forward model combines a core-shell particle with a Gaussian random-field interior, a neural surrogate for the monodisperse SAXS map, and a differentiable layer integrating over particle-size distributions. The surrogate reduces prediction cost by four orders of magnitude, while differentiability enables large-scale multi-start fitting and ensemble identifiability analysis. Applied to synthetic and experimental MC3 LNP data, the framework shows that near-identical SAXS fits can arise from distinct parameter modes, with the experimental fits dominated by a trade-off between size-distribution and interior-structure parameters. △ Less

Submitted 22 May, 2026; originally announced June 2026.

Comments: 38 pages, 24 figures, 5 tables (incl. supplementary information)

GraphAlg Playground: An Online Platform for Learning and Experimenting with the GraphAlg Language

Authors: Daan de Graaf, Robert Brijder, Soham Chakraborty, George Fletcher, Bram van de Wall, Nikolay Yakovets

Abstract: The GraphAlg language for graph algorithms enables native support for user-defined graph analytics workloads in databases. In this demonstration, we present a web-based playground for writing and executing GraphAlg programs in the web browser, including an interactive tutorial explaining its key concepts. The playground runs inside the user's web browser without any installation, and is freely ava… ▽ More The GraphAlg language for graph algorithms enables native support for user-defined graph analytics workloads in databases. In this demonstration, we present a web-based playground for writing and executing GraphAlg programs in the web browser, including an interactive tutorial explaining its key concepts. The playground runs inside the user's web browser without any installation, and is freely available under a permissive license as a reusable library. We present two demonstration scenarios of the publicly available playground website, showing how new users can learn to program in GraphAlg using the tutorial, while expert users can use the playground to prototype and validate their algorithms. △ Less

Submitted 3 June, 2026; originally announced June 2026.

Comments: Accepted at the VLDB 2026 Demonstration Track; to appear in PVLDB Vol. 19. 4 pages, 8 figures, 1 table. Artifacts: https://wildarch.dev/graphalg

Latent Performance Profiling of Large Language Models

Authors: Tanmoy Chakraborty, Ayan Sengupta, Suparna Bhattacharya, Partha Pratim Chakrabarti, Amlan Chakrabarti, Supratik Chakraborty, Partha Pratim Das, Lipika Dey, Richa Singh, Mayank Vatsa

Abstract: Large language models (LLMs) frequently achieve impressive scores on standardized benchmarks, yet accuracy alone offers a limited view of their capabilities. Evaluating open-source LLMs through leaderboards faces persistent issues like data contamination, narrow task scope, and weak alignment with real-world reliability. Benchmark-based evaluations such as MMLU PRO, BBH, or IFEval primarily captur… ▽ More Large language models (LLMs) frequently achieve impressive scores on standardized benchmarks, yet accuracy alone offers a limited view of their capabilities. Evaluating open-source LLMs through leaderboards faces persistent issues like data contamination, narrow task scope, and weak alignment with real-world reliability. Benchmark-based evaluations such as MMLU PRO, BBH, or IFEval primarily capture what a model outputs on fixed test sets, not how it processes information, calibrates uncertainty, or structures internal knowledge. In this article, we advocate for a shift from benchmark-centric evaluation toward a complementary, state-centered intrinsic assessment of LLMs. To this end, we introduce Latent Performance Profiling (LPP) -- a framework that derives task-agnostic diagnostics from hidden activations and output distributions. LPP defines a set of scalar metrics on a model's latent representations and dynamics, revealing scale-independent traits that enable interpretable comparisons and uncover hidden vulnerabilities. Unlike static accuracy scores, LPP provides stable, architecture-sensitive signatures across models of similar size. With extensive empirical analyses across eight LLMs, spanning a size range of 0.5B-14B, we demonstrate that models with similar benchmark scores can exhibit contrasting latent profiles, such as differences in entropy or adaptability. Guided by these insights, we design synthetic probes for uncertainty and symbolic reasoning that align with intrinsic metrics while decoupling from leaderboard bias. We recommend that reporting LPP alongside benchmarks provides a deeper, interpretable understanding of model behavior, enabling more reliable model selection, safety assessment, and evaluation beyond surface-level accuracy. △ Less

Submitted 29 May, 2026; v1 submitted 28 May, 2026; originally announced May 2026.

RL with Learnable Textual Feedback: A Bilevel Approach

Authors: Utsav Singh, Sidhaarth Sredharan, Souradip Chakraborty, Amrit Singh Bedi

Abstract: Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This has motivated a growing line of work on RL with textual feedback, where a critic model generates natural language feedback to guide a reasoning model (the actor), augmenting scalar rewards with richer learning signals. However, existing methods ty… ▽ More Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This has motivated a growing line of work on RL with textual feedback, where a critic model generates natural language feedback to guide a reasoning model (the actor), augmenting scalar rewards with richer learning signals. However, existing methods typically treat feedback as fixed or auxiliary, which misses a key property: feedback should not merely be correct, but should improve the policy (actor model) when provided in context. This motivates a paradigm of learnable textual feedback for RL. Yet the learnability and usefulness of feedback depend on the policy's ability to learn from it, making RL with learnable feedback an inherently bilevel problem. We formalize this coupling as a Stackelberg bilevel program and derive Bilevel Natural Language Actor-Critic (Bi-NAC), which jointly trains a critic to generate reward-improving feedback and an actor to exploit it. Across MATH-500, MBPP, and GPQA, Bi-NAC improves sample and parameter efficiency over RL and fixed-critic baselines: our 2B model outperforms the 3B GRPO baseline, achieving 46.6% versus 41.4% on MATH-500, while our 6B model surpasses the 7B GRPO baseline, achieving 49.3% versus 43.6% on GPQA. △ Less

Submitted 23 May, 2026; originally announced May 2026.

VectorArk: Learning Practical Image Vectorization with Rounded Polygon Representation

Authors: Tarun Gehlaut, Difan Liu, Charu Bansal, Krutik Malani, Souymodip Chakraborty, Ankit Phogat, Matthew Fisher, Vineet Batra

Abstract: Recent vision-language model (VLM)-based approaches have achieved impressive results on image vectorization tasks. However, they are typically evaluated on synthetic benchmarks, where clean SVGs are rasterized at high resolution and then re-vectorized. As a result, these methods generalize poorly to real-world scenarios, such as images with unknown rasterization methods or those generated by text-… ▽ More Recent vision-language model (VLM)-based approaches have achieved impressive results on image vectorization tasks. However, they are typically evaluated on synthetic benchmarks, where clean SVGs are rasterized at high resolution and then re-vectorized. As a result, these methods generalize poorly to real-world scenarios, such as images with unknown rasterization methods or those generated by text-to-image models. We introduce VectorArk, a new VLM-based model designed for robust and practical image vectorization. VectorArk employs a novel rounded polygon representation that simplifies the learning process while naturally producing smooth, visually appealing primitives. We also propose a degradation model that enhances robustness across diverse and imperfect inputs. Our experiments show that, in contrast to previous methods, VectorArk achieves superior geometric completeness and artifact suppression across multiple datasets, with comprehensive ablations validating the contribution of each component. △ Less

Submitted 23 May, 2026; originally announced May 2026.

Comments: CVPR 2026. Project page: https://vectorark.github.io/

Program Synthesis for Non-Linear Real Arithmetic: Going Beyond Realizability

Authors: S. Akshay, Supratik Chakraborty, R. Govind, Aniruddha R. Joshi

Abstract: We study the problem of synthesizing programs from nonlinear real arithmetic (NRA) specifications. Existing techniques, such as syntax-guided synthesis (SyGuS), fail to synthesize programs when the specification is unrealizable. We argue this is unsatisfactory in many situations, and aim to synthesize programs from arbitrary NRA specifications, such that for any input, the synthesized program eith… ▽ More We study the problem of synthesizing programs from nonlinear real arithmetic (NRA) specifications. Existing techniques, such as syntax-guided synthesis (SyGuS), fail to synthesize programs when the specification is unrealizable. We argue this is unsatisfactory in many situations, and aim to synthesize programs from arbitrary NRA specifications, such that for any input, the synthesized program either produces outputs satisfying the specification or reports non-existence of any such output. To avoid rounding errors inherent in floating-point arithmetic, we restrict our programs to work on rational inputs and outputs. We first show that our variant of the synthesis problem is as hard as a long-standing open problem in number theory, and that synthesizing loop-free programs from arbitrary NRA specifications with rational inputs and outputs is impossible in general. Second, we present a sound and complete synthesis algorithm for the case where the specification involves a single output variable. We also show that for realizable specifications, a program generated by SyGuS for NRA (real inputs and outputs) serves as a solution to our problem, where inputs and outputs are rationals. Third, we provide a sound (but necessarily incomplete) synthesis algorithm for the general case of specifications. We have implemented our approach in a prototype tool called NQSynth that solves many benchmarks beyond the reach of state-of-the-art SyGuS tools, even when we render the specifications realizable. △ Less

Submitted 22 May, 2026; originally announced May 2026.

Comparative Analysis of Military Detection Using Drone Imagery Across Multiple Visual Spectrums

Authors: Sourov Roy Shuvo, Prajwal Panth, Rajesh Chowdhury, Sorup Chakraborty, Sudip Chakrabarty, Prasant Kumar Pattnaik

Abstract: In modern warfare, drones are becoming an essential part of intelligence gathering and carrying out precise attacks in different kinds of hostile environments. Their ability to operate in real-time and hostile environments from a safe distance makes them invaluable for surveillance and military operations. The KIIT-MiTA dataset is comprised of images of different military scenarios taken from dron… ▽ More In modern warfare, drones are becoming an essential part of intelligence gathering and carrying out precise attacks in different kinds of hostile environments. Their ability to operate in real-time and hostile environments from a safe distance makes them invaluable for surveillance and military operations. The KIIT-MiTA dataset is comprised of images of different military scenarios taken from drones, and these provide a foundation for detecting military objects, but it does not take into account the various types of real-world scenarios. With that in mind, to evaluate how the models are performing under varying conditions, four different types of datasets are created: Gray Scale, Thermal Vision, Night Vision, and Obscura Vision. These simulate the real-world environments such as low visibility, heat-based imagery, and nighttime conditions. The YOLOv11-small model is trained and used to detect objects across diverse settings. This research boosts the performance and reliability of drone-based operations by contributing to the development of advanced detection systems in both defensive and offensive missions. △ Less

Submitted 20 May, 2026; originally announced May 2026.

Comments: 6 pages, 7 figures. Accepted at the 16th International Conference on Computing, Communication and Networking Technologies (ICCCNT), July 6-11, 2025, IIT Indore. Proceedings pending publication

Fast and Featureless Node Representation Learning with Partial Pairwise Supervision

Authors: Sujan Chakraborty, Saptarshi Bej

Abstract: We introduce Contrastive FUSE, a fast and unified framework for scalable node representation learning in graphs with partially available pairwise node labels and no available node features. Unlike existing methods, we directly optimize a spectral contrastive objective that integrates community-aware structural signals with signed pairwise constraints. To support large-scale training, we replace th… ▽ More We introduce Contrastive FUSE, a fast and unified framework for scalable node representation learning in graphs with partially available pairwise node labels and no available node features. Unlike existing methods, we directly optimize a spectral contrastive objective that integrates community-aware structural signals with signed pairwise constraints. To support large-scale training, we replace the expensive modularity gradient with a lightweight approximation, which preserves the structure-seeking behavior of modularity while reducing the computational cost significantly. This yields an efficient optimization scheme with a natural gradient decomposition and adaptive learning-rate scaling, enabling fast iterative updates even on million-edge graphs. Extensive experiments on benchmark citation networks, large co-purchase graphs, and OGB datasets show that Contrastive FUSE achieves competitive or superior contrastive classification performance without relying on node features, while offering substantial runtime gains over existing baselines. These results highlight the effectiveness of coupling modularity-inspired structural learning with contrastive supervision for efficient and scalable contrastive node representation learning. △ Less

Submitted 19 May, 2026; originally announced May 2026.

ASTRA: Asynchronous Age-Aware Satellite Random Access via Mean-Field Control

Authors: Sayam Chakraborty, Aimin Li, Yigit Ince, Sajjad Baghaee, Elif Uysal

Abstract: Satellite Internet-of-Things (IoT) enables massive status-update services beyond terrestrial coverage, but grant-free uplink access creates a coupled freshness-control problem: increasing repetition and receiver-side diversity improves a device's capture-SIC opportunities, yet the resulting population congestion degrades network-wide freshness. Existing AoI-aware random-access models often rely on… ▽ More Satellite Internet-of-Things (IoT) enables massive status-update services beyond terrestrial coverage, but grant-free uplink access creates a coupled freshness-control problem: increasing repetition and receiver-side diversity improves a device's capture-SIC opportunities, yet the resulting population congestion degrades network-wide freshness. Existing AoI-aware random-access models often rely on slot-synchronous collisions, fixed delivery probabilities, or scalar transmit-or-wait decisions and therefore cannot capture asynchronous satellite uplinks with capture and SIC. This paper develops a PHY-aware mean-field framework, termed ASTRA (Asynchronous Age-Aware Satellite Random Access), for freshness-driven satellite IoT random access. We build an access model that captures asynchronous arrivals, partial overlaps, capture, and SIC while preserving the dependence of delivery success on each device's repetition-diversity action. We then formulate the population interaction as a scalable mean-field MDP in which devices optimize access timing and intensity using only local AoI observations. The resulting system admits a mean-field equilibrium in which individual optimality and endogenous congestion are mutually consistent. We further prove that the optimal equilibrium policy admits an age-threshold structure. Numerical results show that the proposed policy reduces AoI relative to age-independent baselines. △ Less

Submitted 18 May, 2026; originally announced May 2026.

3C: Competition, Competence, and Collaboration for Women in Computing

Authors: Ioana Visescu, Shalini Chakraborty

Abstract: Women in computer science and software engineering continue to face structural and cultural barriers affecting recognition, collaboration, and career progression. Existing environments often reinforce competition, tokenism, and exclusion, particularly in male dominated academic and professional spaces. This extended abstract introduces the 3C framework Competition, Competence, and Collaboration to… ▽ More Women in computer science and software engineering continue to face structural and cultural barriers affecting recognition, collaboration, and career progression. Existing environments often reinforce competition, tokenism, and exclusion, particularly in male dominated academic and professional spaces. This extended abstract introduces the 3C framework Competition, Competence, and Collaboration to explore how women experience and navigate networking in computing environments. We discuss how perceptions of competence, access to collaborative networks, and competition for limited opportunities shape womens' participation and sense of belonging. As a call to action, we propose community driven discussions, focus groups, and participatory data collection within the ACM womENcourage community to better understand and address these challenges. Our goal is to foster stronger networks of mentorship, solidarity, and collaboration among women in computing. △ Less

Submitted 13 May, 2026; originally announced May 2026.

GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?

Authors: Kaixiang Zhao, Bolin Shen, Yuyang Dai, Shayok Chakraborty, Yushun Dong

Abstract: Graph neural networks (GNNs) deployed as cloud services can be stolen through model-extraction attacks, which train a surrogate from query responses to reproduce the target's behavior, and a growing line of ownership defenses tries to prevent or trace such theft. This paper asks two questions: how hard is it to steal a GNN, and can we stop it? Prior work cannot answer either, because experiments u… ▽ More Graph neural networks (GNNs) deployed as cloud services can be stolen through model-extraction attacks, which train a surrogate from query responses to reproduce the target's behavior, and a growing line of ownership defenses tries to prevent or trace such theft. This paper asks two questions: how hard is it to steal a GNN, and can we stop it? Prior work cannot answer either, because experiments use inconsistent datasets, threat models, and metrics. We introduce GraphIP-Bench, a unified benchmark that evaluates both sides under a single black-box protocol. GraphIP-Bench integrates twelve extraction attacks, twelve defenses spanning watermarking, output perturbation, and query-pattern detection, ten public graphs covering homophilic, heterophilic, and large-scale regimes, three GNN backbones, and three graph-learning tasks. It reports fidelity, task utility, ownership verification, and computational cost on shared splits, queries, and budgets. We further add a joint attack-and-defense track that runs every attack on every defended target and measures watermark verification on the resulting surrogate, exposing how much protection a defense retains after extraction. The empirical picture is clear: stealing a GNN is easy at medium query budgets and most defenses do not change this; several watermarks verify reliably on the protected model but lose most of their verification signal on the extracted surrogate, exposing a gap that single-model evaluations miss; and heterophilic graphs are systematically harder to steal, while a cross-architecture mismatch between target and surrogate reduces but does not prevent extraction. We release GraphIP-Bench with reproducible scripts and configurations, and integrate the attacks and defenses into the PyGIP library. Code: https://github.com/LabRAI/GraphIP-Bench. Library: https://labrai.github.io/PyGIP/index.html. △ Less

Submitted 25 May, 2026; v1 submitted 12 May, 2026; originally announced May 2026.

Comments: Under review

CoT-Guard: Small Models for Strong Monitoring

Authors: Nirav Diwan, Han Wang, Berkcan Kapusuzoglu, Ramin Moradi, Supriyo Chakraborty, Giri Iyengar, Sambit Sahu, Huan Zhang, Gang Wang

Abstract: Monitoring the chain-of-thought (CoT) of reasoning models is a promising approach for detecting covert misbehavior (i.e., hidden objectives) in code generation tasks. While large models (GPT-5, Gemini-3-Flash) can serve as effective CoT monitors, they are expensive to deploy due to the lengthy reasoning traces and high API cost, emphasizing the need for smaller, cheaper alternatives. Nevertheless,… ▽ More Monitoring the chain-of-thought (CoT) of reasoning models is a promising approach for detecting covert misbehavior (i.e., hidden objectives) in code generation tasks. While large models (GPT-5, Gemini-3-Flash) can serve as effective CoT monitors, they are expensive to deploy due to the lengthy reasoning traces and high API cost, emphasizing the need for smaller, cheaper alternatives. Nevertheless, we find that current small models (4B--8B) struggle to detect hidden objectives despite access to the CoT, frequently misattributing them as part of the user query. To address this, we propose a post-training pipeline combining supervised fine-tuning (SFT) and reinforcement learning (RL), where SFT narrows the gap for in-domain tasks by distilling detection behavior from stronger monitors, and RL on hard and subtly crafted hidden objectives helps the model generalize to out-of-domain monitoring tasks. To validate this generalization, we evaluate under a realistic threat model motivated by practical supply-chain attacks, where the adversary is a third-party LLM router injecting hidden objectives into code-generation requests through either prompt manipulation or code manipulation attacks. To push beyond objectives that large monitors already saturate, we also introduce four new challenging tasks even for strong monitors. Finally, we introduce CoT-Guard, a 4B-parameter monitor that demonstrates superior generalization performance under both prompt and code manipulation attacks, achieving a G-mean^2 (i.e., TNR x TPR) of 75% and outperforming GPT-5.4 (56%), GPT-5-mini (41%), and Qwen3-32B (54%), while closing the gap to Gemini-3-Flash (83%). These results demonstrate that CoT-Guard provides a practical and cost-effective user-side defense, substantially improving hidden-objective detection while avoiding the deployment cost of large monitors. △ Less

Submitted 12 May, 2026; originally announced May 2026.

Stochastic Smoothed Particle Hydrodynamics for Stochastic Mechanics Problems

Authors: Mridul Tiwari, Sawan Kumar, Md Rushdie Ibne Islam, Souvik Chakraborty

Abstract: Smoothed Particle Hydrodynamics (SPH_ is a mesh-free Lagrangian method renowned for modeling large deformations and free-surface flows, yet classical formulations remain confined to deterministic systems. We introduce Stochastic SPH (S-SPH), which employs orthogonal Polynomial Chaos expansions to represent uncertainties in system parameters, forcing functions, and boundary or initial conditions, w… ▽ More Smoothed Particle Hydrodynamics (SPH_ is a mesh-free Lagrangian method renowned for modeling large deformations and free-surface flows, yet classical formulations remain confined to deterministic systems. We introduce Stochastic SPH (S-SPH), which employs orthogonal Polynomial Chaos expansions to represent uncertainties in system parameters, forcing functions, and boundary or initial conditions, while spatial variation is captured via the SPH kernel. Random fields are discretized through Karhunen-Loève expansions, and a Galerkin projection in the polynomial basis transforms the underlying SPDE into a coupled system of ordinary differential equations governing the time evolution of expansion coefficients. To enforce Dirichlet and Neumann conditions in a mesh-free context, ghost-particle techniques augmented by a gradient-correction matrix are employed, and a predictor-corrector integration scheme ensures numerical stability. We validate S-SPH on benchmark problems, including one-dimensional advection with stochastic advection speed, inviscid Burgers' equations with random initial amplitudes, and two-dimensional Burgers' flows with uncertain Fourier-mode initial fields and viscosity, demonstrating excellent agreement with Monte Carlo simulation statistics of mean and variance. Remarkably, S-SPH achieves up to three orders of magnitude reduction in computational cost relative to direct sampling approaches. The proposed framework thus provides an efficient, accurate, and fully mesh-free methodology for uncertainty quantification in complex mechanics applications. △ Less

Submitted 7 May, 2026; originally announced May 2026.

Comments: 23 pages, 11 figures

PoHAR: Understanding Hyperlocal Human Activities with Pollution Sensor Networks

Authors: Prasenjit Karmakar, Karthik Reddy, Sandip Chakraborty

Abstract: Low-cost air quality sensors are becoming ubiquitous in our daily lives as public awareness of air pollution continues to grow, and people take measures to monitor and improve the air they breathe indoors. Besides the standard operation of these sensors, fluctuations in environmental parameters can be leveraged to understand human behavior and activities in indoor spaces. Unlike traditional audio-… ▽ More Low-cost air quality sensors are becoming ubiquitous in our daily lives as public awareness of air pollution continues to grow, and people take measures to monitor and improve the air they breathe indoors. Besides the standard operation of these sensors, fluctuations in environmental parameters can be leveraged to understand human behavior and activities in indoor spaces. Unlike traditional audio-visual, Radio Frequency, and inertial sensors, air quality sensors are easily scalable to a household, are privacy-preserving, and more economical. Such distributed sensor networks must jointly make decisions to monitor indoor occupants for downstream smart home and healthcare applications. However, due to low processing power, memory, and energy, they often struggle to maintain distributed data consensus and identify activity-affected sensor groups for accurate on-device inference. In this paper, we propose PoHAR framework that implements: (i) a conflict-free replicated data primitive for data sharing, (ii) a hierarchical clustering for ESP32 to detect activity-affected sensor groups with a self-supervised distance metric, and (iii) a leader-based group inference with off-the-shelf ML classifiers, enabling the sensor network to collaboratively detect hyperlocal indoor activities. Our extensive experiments demonstrated on-device activity detection, achieving 97.41% accuracy for indoor activity and 99.68% for cooking activity, using off-the-shelf ML models with latency below 34 microseconds. △ Less

Submitted 10 May, 2026; originally announced May 2026.

Comments: 8 pages, 8 figures, accepted to IEEE DCOSS-IoT 2026

Recursive Agent Optimization

Authors: Apurva Gandhi, Satyaki Chakraborty, Xiangjun Wang, Aviral Kumar, Graham Neubig

Abstract: We introduce Recursive Agent Optimization (RAO), a reinforcement learning approach for training recursive agents: agents that can spawn and delegate sub-tasks to new instantiations of themselves recursively. Recursive agents implement an inference-time scaling algorithm that naturally allows agents to scale to longer contexts and generalize to more difficult problems via divide-and-conquer. RAO pr… ▽ More We introduce Recursive Agent Optimization (RAO), a reinforcement learning approach for training recursive agents: agents that can spawn and delegate sub-tasks to new instantiations of themselves recursively. Recursive agents implement an inference-time scaling algorithm that naturally allows agents to scale to longer contexts and generalize to more difficult problems via divide-and-conquer. RAO provides a method to train models to best take advantage of such recursive inference, teaching agents when and how to delegate and communicate. We find that recursive agents trained in this way enjoy better training efficiency, can scale to tasks that go beyond the model's context window, generalize to tasks much harder than the ones the agent was trained on, and can enjoy reduced wall-clock time compared to single-agent systems. △ Less

Submitted 7 May, 2026; originally announced May 2026.

Breaking In and Reaching Out: Networking for Women in Computer Science

Authors: Shalini Chakraborty

Abstract: Networking is central to careers in computer science, where a globally distributed and diverse community increasingly collaborates across institutional and geographic boundaries, often in hybrid and remote settings. However, access to effective networking is shaped by structural and personal factors, including geography, funding, language, identity, personality, and caregiving responsibilities. Bu… ▽ More Networking is central to careers in computer science, where a globally distributed and diverse community increasingly collaborates across institutional and geographic boundaries, often in hybrid and remote settings. However, access to effective networking is shaped by structural and personal factors, including geography, funding, language, identity, personality, and caregiving responsibilities. Building on prior work, this workshop focuses on women in computing to examine lived experiences of networking and the barriers they encounter. Through a community-driven discussion grounded in a factor-based framework, the workshop aims to surface overlooked challenges and foster shared understanding. Ultimately, it seeks to inform more inclusive, equitable, and accessible networking practices within the computer science community. △ Less

Submitted 7 May, 2026; originally announced May 2026.

Knowledge Compilation for Quantification in Alternating Automata

Authors: S. Akshay, Alfredo Cantarella, Supratik Chakraborty, Bernd Finkbeiner, Niklas Metzger

Abstract: We present a knowledge compilation approach for existential and universal quantification in alternating automata. Knowledge compilation transforms formulas into normal forms with special properties that enable efficient answering of questions of interest. For Boolean formulas, several normal forms that have proven effective for existential/universal quantification, and even for functional synthesi… ▽ More We present a knowledge compilation approach for existential and universal quantification in alternating automata. Knowledge compilation transforms formulas into normal forms with special properties that enable efficient answering of questions of interest. For Boolean formulas, several normal forms that have proven effective for existential/universal quantification, and even for functional synthesis, have been studied in the literature. For infinite word automata, quantification is a fundamental operation in verification tasks such as QPTL satisfiability checking and HyperLTL model checking. Existing algorithms rely on nondeterministic infinite word automata, where existential projection can be efficiently performed state-wise, but universal projection requires complementation. Complementing nondeterministic infinite word automata, however, is expensive in practice, making existing algorithms infeasible for automata in practice. Towards addressing this problem, we propose novel knowledge compilation techniques for existential and universal quantification on alternating safety automata. Our approach compiles alternating automata into normal forms where projection can be applied uniformly and efficiently to each state's transition function. Using the compilations for each type of quantification, we can effectively eliminate a sequence of alternating quantifiers in formulas without complementation. Our BDD-based prototype demonstrates the practical effectiveness of our algorithms on a suite of QPTL satisfiability benchmarks. △ Less

Submitted 3 May, 2026; originally announced May 2026.

Comments: Published at the 23rd International Conference on Principles of Knowledge Representation and Reasoning

Autonomous Systems Dependability in the era of AI: Design Challenges in Safety, Security, Reliability and Certification

Authors: Behnaz Ranjbar, Kirankumar Raveendiran, Sudeep Pasricha, Samarjit Chakraborty, Cecilia Carbonelli, Akash Kumar

Abstract: The design of embedded safety-critical systems such as those used in next-generation automotive and autonomous platforms, is increasingly challenged by escalating system complexity, hardware-software heterogeneity, and the integration of intelligent, data-driven components. Ensuring dependability in such systems requires a holistic approach that spans multiple abstraction layers and encompasses bo… ▽ More The design of embedded safety-critical systems such as those used in next-generation automotive and autonomous platforms, is increasingly challenged by escalating system complexity, hardware-software heterogeneity, and the integration of intelligent, data-driven components. Ensuring dependability in such systems requires a holistic approach that spans multiple abstraction layers and encompasses both design- and run-time assurance. Traditional methods for reliability, safety, and security management often fall short in addressing the dynamic and uncertain behaviors introduced by Artificial Intelligence (AI) and Machine Learning (ML) components, especially under stringent real-time, power, and safety constraints. While AI and ML offer powerful predictive, adaptive, and self-optimizing capabilities that can enhance system dependability, their inherent non-determinism, data-dependence, and lack of formal guarantees introduce new challenges for verification, validation, and certification. This paper explores emerging methodologies, architectures, and frameworks for designing dependable autonomous and embedded systems in the era of AI. It highlight advances in reliability modeling, secure system design, and certification approaches that account for imperfect, learning-enabled components, aiming to bridge the gap between AI innovation and certifiable system-level dependability. △ Less

Submitted 1 May, 2026; v1 submitted 30 April, 2026; originally announced April 2026.

On-the-fly LTLf Synthesis under Partial Observability

Authors: Nadav Alon, Supratik Chakraborty, Alexandre Duret-Lutz, Dror Fried, Lucas M. Tabajara, Moshe Y. Vardi, Shufang Zhu

Abstract: LTLf synthesis under partial observability requires reasoning about unobservable environment variables, which is typically handled by constructing a belief-state DFA via subset construction that universally quantifies these variables. Existing approaches perform this construction as a separate step prior to game solving, often generating belief states that are unnecessary in practice. We propose a… ▽ More LTLf synthesis under partial observability requires reasoning about unobservable environment variables, which is typically handled by constructing a belief-state DFA via subset construction that universally quantifies these variables. Existing approaches perform this construction as a separate step prior to game solving, often generating belief states that are unnecessary in practice. We propose an on-the-fly approach to LTLf synthesis under partial observability based on observable progression. Our method incrementally builds the belief-state DFA by progressing the specification with respect to observable variables only, universally quantifying unobservable variables on the fly. We prove the correctness of the construction and show that it naturally enables on-the-fly game solving, leading to a fully on-the-fly synthesis framework. Our implementation leverages DFAs represented using Multi-Terminal Binary Decision Diagrams: a compact representation that has proven highly effective for LTLf synthesis under full observability. Experimental results demonstrate that our approach significantly outperforms existing methods and further highlight the practical benefits of integrating on-the-fly game solving with belief-state construction. △ Less

Submitted 3 May, 2026; v1 submitted 29 April, 2026; originally announced April 2026.

Comments: To appear in Proceedings of the 26th International Conference on Principles of Knowledge Representation and Reasoning (KR2026), 9 pages + references and appendix

Maximum Matching and Related Problems in Catalytic Logspace

Authors: Srijan Chakraborty, Samir Datta, Aryan Kusre, Partha Mukhopadhyay, Amit Sinhababu

Abstract: Understanding the power of space-bounded computation with access to catalytic space has been an important theme in complexity theory over the recent years. One of the key algorithmic results in this area is that bipartite maximum matching can be computed in catalytic logspace with a polynomial-time bound, Agarwala and Mertz (2025). In this paper, we show that we can construct a \emph{maximum mat… ▽ More Understanding the power of space-bounded computation with access to catalytic space has been an important theme in complexity theory over the recent years. One of the key algorithmic results in this area is that bipartite maximum matching can be computed in catalytic logspace with a polynomial-time bound, Agarwala and Mertz (2025). In this paper, we show that we can construct a \emph{maximum matching} in \emph{general graphs} in CL, and, in fact, in CLP. We first show that the size of a \emph{maximum matching} in \emph{general graphs} can be determined in CL. Our algorithm is based on the linear-algebraic algorithm for maximum matching by Geelen (2000). We then show that this algorithm, along with some new ideas, can be used to \emph{find} a maximum matching in general graphs. Using a similar algorithm of Geelen (1999), we also solve the \emph{maximum rank completion problem} in CLP, which was previously known to be solvable in deterministic polynomial time, Geelen. This problem turns out to be equivalent to the \emph{linear matroid intersection} problem (shown by Murota, 1995) which has been shown to be in CLP by Agarwala, Alekseev, and Vinciguerra (2026). Finally, using a PTAS algorithm Bläser, Jindal and Pandey (2018), for approximating the rank in Edmond's problem, we derive a CLP algorithm that can approximate the rank given by any instance of the \emph{Edmond's problem} upto a factor of $(1-\eps)$ for any $\eps\in(0,1)$. An application of this is a CLP bound for approximating the maximum independent matching size in the \emph{linear matroid matching} problem. △ Less

Submitted 31 May, 2026; v1 submitted 27 April, 2026; originally announced April 2026.

Comments: Preliminary version