arXiv subjects

Swastik Roy

Publications and source records attributed to Swastik Roy.

7 recordsLinked to original sources

arXiv2026

AGC-Bench: Measuring Artificial General Creativity

Creativity research has debated whether creativity is domain-specific (e.g., visual, writing, science), and if it is psychometrically separable from general intelligence. Both questions now apply to LLMs, but a unified benchmark of AI creativity remains elusive. We introduce AGC-Bench, an artificial general creativity benchmark built from a systematic review of the AI creativity literature (3,101 papers screened, 497 benchmarks identified), paired with an agentic harness that converts idiosyncratic codebases into HELM-standardized benchmarks. The first release covers 78 datasets spanning brainstorming, problem solving, STEM, narrative, figurative language, and humor. To address bias in LLM-as-judge, we apply Judge Response Theory -- a psychometric calibration of judge leniency/severity; we then fine-tune Qwen3-30B on the bias-corrected ratings of three frontier LLMs to produce AGC-Judge, an open-weight model that robustly scores new creativity benchmarks it was not trained on. Results reveal frontier models at the top of the AGC-Bench leaderboard, with open models close behind. LLMs show different creative strengths, ranking higher on some domains (e.g., writing) than others (e.g., scientific ideation). Extensive experiments yield three main findings. First, applying factor analysis across 83 LLMs, we recover a single creativity factor 'c', analogous to the 'g' factor of general intelligence, that explains 81.5% of variance, related to but separable from general knowledge/reasoning. Second, we show that prompting models to "be creative" boosts their performance far more than enabling reasoning, evidence that the benchmark tracks creativity over general ability. Third, on a human-matched subset, we find the top human still leads the top LLM on creativity. We release AGC-Bench with a public leaderboard, AGC-Judge, and human data as open infrastructure for measuring AI creativity at scale.

cs.CL↗

arXiv2026

PReMISE: Policy Rubrics as Measurement Specifications for LLM Judges

LLM judges are increasingly used to evaluate open-ended responses, but their scores depend strongly on the rubrics that condition them. A vague rubric asking for a response to be ``helpful and factual'' can reward polished answers that invent facts or violate user intent. We treat reusable rubrics as measurement specifications: changing the rubric changes the response quality measurement induced by a fixed judge. We introduce PReMISE, a framework that, given pairwise human-preference data, (i) discovers a policy-level rubric set, and (ii) audits any rubric set under LLM-judge use along four axes: structural adequacy, reliability, preference fit, and adversarial robustness. Across rubric sources no raw source is simultaneously reliable, preference-predictive, and adversarially robust; and high inter-rater agreement does not imply low exploitability. PReMISE is the only rubric source to score non-trivially on applicability, specificity, and effective dimensionality simultaneously. We contribute two audit-targeted repair operations: preference-rank selection raises judge accuracy on paired responses from $65.0\%$ to $68.6\%$, competitive with the strongest rubric-discovery baselines and leading on two of three judges in our cross-judge sweep; reliability-constrained refinement reduces the rate at which exploit responses receive high scores from $46.4\%$ to $36.0\%$ with little change in inter-judge agreement ($\alpha{=}.531\to.519$).

cs.AI↗

arXiv2026

C-MORAL: Controllable Multi-Objective Molecular Optimization with Reinforcement Alignment for LLMs

Large language models (LLMs) show promise for molecular optimization, but aligning them with selective and competing drug-design constraints remains challenging. We propose C-Moral, a reinforcement learning post-training framework for controllable multi-objective molecular optimization. C-Moral combines group-based relative optimization, property score alignment for heterogeneous objectives, and bottleneck-sensitive non-linear reward aggregation to improve stability across competing molecular properties. Experiments on C-MuMOInstruct and S$^2$-Bench MolOpt show that C-Moral achieves the best performance among compared methods on both benchmarks. On C-MuMOInstruct, C-Moral achieves the best Success Optimized Rate (SOR) of 48.9\% on in-domain tasks and 39.5\% on out-of-domain tasks while preserving scaffold similarity. On S$^2$-Bench MolOpt, it also achieves the strongest results across LogP, MR, and QED optimization tasks. These results suggest that C-Moral is an effective way to align molecular LLMs with continuous and constrained molecular design objectives. Our code and models are publicly available at https://github.com/Rwigie/C-MORAL.

cs.LG↗

arXiv2025

ScalePRM: Training Process Reward Models by Scaling Verification Compute Without Ground Truth

Training process reward models (PRMs) requires step-level correctness labels, obtained either through expensive human annotation or by relying on ground-truth answers, limiting the ability to scale process-level supervision. We propose ScalePRM, which scales verification compute as an alternative: given a problem and a candidate solution, we generate multiple independent verifications of each reasoning step and aggregate their judgments to produce synthetic step-level labels without ground truth. We explore two representative inference-time scaling strategies, parallel scaling through self-consistency and sequential scaling through meta-critique, and train generative PRMs on the resulting synthetic data. On ProcessBench, a benchmark for identifying erroneous steps in mathematical reasoning, PRMs trained on step-level self-consistency data achieve 67.5 F1, surpassing reference-guided training with ground-truth access (66.4 F1) and GPT-4o as a critic (61.9 F1). When deployed as reward signals in RL training with Qwen2.5-Math-7B, our best PRM achieves 47.4% average accuracy across six mathematical reasoning benchmarks, outperforming ground-truth-based RLVR (43.9%). We also identify and address reward exploitation patterns unique to generative PRM-based RL. Our results demonstrate that scaling verification compute is a viable alternative to ground-truth supervision for training process reward models.

cs.LG↗

arXiv2025

JudgeBoard: Benchmarking and Enhancing Small Language Models for Reasoning Evaluation

While small language models (SLMs) have shown promise on various reasoning tasks, their ability to judge the correctness of answers remains unclear compared to large language models (LLMs). Prior work on LLM-as-a-judge frameworks typically relies on comparing candidate answers against ground-truth labels or other candidate answers using predefined metrics like entailment. However, this approach is inherently indirect and difficult to fully automate, offering limited support for fine-grained and scalable evaluation of reasoning outputs. In this work, we propose JudgeBoard, a novel evaluation pipeline that directly queries models to assess the correctness of candidate answers without requiring extra answer comparisons. We focus on two core reasoning domains: mathematical reasoning and science/commonsense reasoning, and construct task-specific evaluation leaderboards using both accuracy-based ranking and an Elo-based rating system across five benchmark datasets, enabling consistent model comparison as judges rather than comparators. To improve judgment performance in lightweight models, we propose MAJ (Multi-Agent Judging), a novel multi-agent evaluation framework that leverages multiple interacting SLMs with distinct reasoning profiles to approximate LLM-level judgment accuracy through collaborative deliberation. Experimental results reveal a significant performance gap between SLMs and LLMs in isolated judging tasks. However, our MAJ framework substantially improves the reliability and consistency of SLMs. On the MATH dataset, MAJ using smaller-sized models as backbones performs comparatively well or even better than their larger-sized counterparts. Our findings highlight that multi-agent SLM systems can potentially match or exceed LLM performance in judgment tasks, with implications for scalable and efficient assessment.

cs.AI↗

arXiv2025

OPTAGENT: Optimizing Multi-Agent LLM Interactions Through Verbal Reinforcement Learning for Enhanced Reasoning

Large Language Models (LLMs) have shown remarkable reasoning capabilities in mathematical and scientific tasks. To enhance complex reasoning, multi-agent systems have been proposed to harness the collective intelligence of LLM agents. However, existing collaboration structures are either predefined or rely on majority voting or round-table debates, which can suppress correct but less dominant agent contributions. Recent approaches model multi-agent systems as graph networks but optimize purely for agent performance, neglecting the quality of interactions. We hypothesize that effective agent communication is crucial for multi-agent reasoning and that debating quality plays a significant role. To address this, we propose $\ours$, a multi-agent verbal reinforcement learning algorithm that dynamically constructs and refines multi-agent collaboration structures. Our method defines action spaces and a feedback mechanism that evaluates communication robustness and coherence throughout the debate. The final decision is achieved through a majority vote over all the agents. We assess $\ours$ on various reasoning tasks, including mathematical reasoning, creative writing, scientific reasoning, and numerical sorting. Results demonstrate that our approach significantly outperforms single-agent prompting methods and state-of-the-art multi-agent frameworks on diverse tasks.

cs.AI↗

arXiv2025

BeyondBench: Contamination-Resistant Evaluation of Reasoning in Language Models

Evaluating language models fairly is increasingly difficult as static benchmarks risk contamination by training data, obscuring whether models truly reason or recall. We introduce BeyondBench, an evaluation framework using algorithmic problem generation to create mathematically grounded problems on the fly, ensuring each test remains uncontaminated. Our framework covers 44 algorithmic tasks with 117 variations across three difficulty levels: the Easy Suite (29 tasks) for arithmetic and statistics, the Medium Suite (5 tasks, 49 variations) for sequence patterns and reasoning, and the Hard Suite (10 tasks, 68 variations) for NP-complete and constraint satisfaction problems. Each task draws from a space exceeding 10^15 unique instances, with deterministically verified solutions. We evaluated 101 language models (85 open-source, 16 closed-source), spanning 0.5B to 141B parameters and multiple quantization schemes, using three-fold evaluation for robustness. Results reveal consistent reasoning deficiencies, with performance degrading sharply as complexity increases. In Hard Suite evaluations, Gemini-2.5-pro, Llama-3.3-70B, and Qwen2.5-72B achieved accuracies of 56.21%, 27.16%, and 33.37% respectively. Performance drops significantly without tool usage, with GPT-5, GPT-5-mini, and GPT-5-nano showing declines of 16.81%, 15.86%, and 43.95% in overall accuracy. Contamination resistance rests on three guarantees: (i) the problem space vastly exceeds any static dataset, (ii) every instance has a deterministically verifiable solution, and (iii) isomorphic transformations yield semantically equivalent but syntactically novel problems. BeyondBench redefines reasoning evaluation via genuine algorithmic problem-solving. Our leaderboard is at https://ctrl-gaurav.github.io/BeyondBench/, Python package at https://pypi.org/project/beyondbench/, and codebase at https://github.com/ctrl-gaurav/BeyondBench.

cs.CL↗