Subramanyam Sahoo, Aman Chadha, Vinija Jain, Divya Chaudhary · Mar 6, 2026 · Citations: 0
Critique EditMathCoding
We introduce SAHOO, a practical framework to monitor and control drift through three safeguards: (i) the Goal Drift Index (GDI), a learned multi-signal detector combining semantic, lexical, structural, and distributional measures; (ii)…
Xisen Jin, Michael Duan, Qin Lin, Aaron Chan, Zhenglun Chen, Junyi Du · Mar 6, 2026 · Citations: 0
Red TeamAutomatic MetricsMathLaw
As AI agents become widely deployed as online services, users often rely on an agent developer's claim about how safety is enforced, which introduces a threat where safety measures are falsely advertised.
To generate proof, the developer runs the agent and guardrail inside a Trusted Execution Environment (TEE), which produces a TEE-signed attestation of guardrail code execution verifiable by any user offline.
We demonstrate the success of replay in practice for fine-tuning 8B parameter models, improving agentic web navigation success by 4.5\% and Basque question-answering accuracy by 2\%.
We formalize this as Theory-Aligned and Distribution-Controllable Persona Generation (TAD-PG) and introduce HACHIMI, a multi-agent Propose-Validate-Revise framework that generates theory-aligned, quota-controlled personas.
Intrinsic evaluation shows near-perfect schema validity, accurate quotas, and substantial diversity, while external evaluation instantiates personas as student agents answering CEPS and PISA 2022 surveys; across 16 cohorts, math and…
On code generation (LiveCodeBench, CodeContests, SWE-Bench) and math reasoning (AIME, HMMT) benchmarks, V_1-Infer improves Pass@1 by up to 10% over pointwise verification and outperforms recent test-time scaling methods while being…
Jiale Lao, Immanuel Trummer · Mar 2, 2026 · Citations: 0
Automatic MetricsMathCoding
As a proof of concept, we present GenDB, an LLM-powered agentic system that generates instance-optimized and customized query execution code tailored to specific data, workloads, and hardware resources.
We implemented an early prototype of GenDB that uses Claude Code Agent as the underlying component in the multi-agent system, and we evaluate it on OLAP workloads.
Aleksei Dorkin, Taido Purason, Emil Kalbaliyev, Hele-Andra Kuulmets, Marii Ojastu, Mark Fišel · Mar 2, 2026 · Citations: 0
Pairwise PreferenceMathCoding
We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior.
Evaluation on a comprehensive suite of Estonian benchmarks shows consistent gains in linguistic competence, knowledge, reasoning, translation quality, and instruction-following compared to the original base model and its instruction-tuned…
While prior research emphasizes the role of on-policy data in mitigating forgetting, we uncover--and validate both theoretically and empirically--an overlooked yet critical mechanism: the implicit regularization inherent in Direct…
Frédéric Berdoz, Yann Billeter, Yann Vonlanthen, Roger Wattenhofer · Mar 1, 2026 · Citations: 0
Pairwise PreferenceMath
Opinion modeling aims to capture individual or group political preferences, enabling applications such as digital democracies, where models could help shape fairer and more popular policies.
We propose AgentDropoutV2 (ADv2), a test-time rectify-or-reject pruning framework that dynamically optimizes MAS information flow.
Empirical results demonstrate that ADv2 significantly boosts performance on both fixed and dynamic MAS frameworks, achieving average accuracy gains of 6.39 and 2.28 percentage points on extensive math and code benchmarks, respectively.
Roy Miles, Aysim Toker, Andreea-Maria Oncescu, Songcen Xu, Jiankang Deng, Ismail Elezi · Feb 26, 2026 · Citations: 0
Automatic MetricsMathCoding
This modular pipeline separates exploration (diffusion) from evaluation and solution synthesis, avoiding monolithic unified hybrids while preserving broad search.
Across math reasoning benchmarks, we find that step-level recombination is most beneficial on harder problems, and ablations highlight the importance of the final AR solver in converting stitched but imperfect rationales into accurate…
We provide empirical evidence through a detailed case study: the discovery of novel error representations and bounds for Hermite quadrature rules via systematic human-AI collaboration.
However, every step required rigorous human verification, mathematical intuition for problem formulation, and strategic direction.
Sweta Karlekar, Carolina Zheng, Magnus Saebo, Nicolas Beltran-Velez, Shuyang Yu, John Bowlan · Feb 25, 2026 · Citations: 0
Pairwise PreferenceAutomatic MetricsMath
Building on this observation, we introduce Duel-Evolve, an evolutionary optimization algorithm that replaces external scalar rewards with pairwise preferences elicited from the same LLM used to generate candidates.
Results show that pairwise self-preferences provide strong optimization signal for test-time improvement over large, discrete output spaces.
Alex Stein, Furong Huang, Tom Goldstein · Feb 24, 2026 · Citations: 0
Automatic MetricsMath
Held-out in-domain accuracy under asymmetric evaluation improves from 46.0\% to 62.0\%, and average (maj@8) accuracy on public document-free math benchmarks improves from 20.2\% to 35.4\%.
Abraham Paul Elenjical, Vivek Hruday Kavuri, Vasudeva Varma · Feb 21, 2026 · Citations: 0
Pairwise PreferenceHuman EvalMathMedicine
We introduce a psychologically grounded metacognitive framework that operationalizes Ann Brown's regulatory cycle (Planning, Monitoring, and Evaluation) as a structured prompting architecture, and study its integration within a lightweight…
Blinded human evaluations over 580 query pairs show an 84% aggregate preference for trustworthiness and metacognitive self-awareness over standard and Chain-of-Thought baselines.
Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, Masashi Sugiyama · Feb 20, 2026 · Citations: 0
Llm As JudgeAutomatic MetricsMath
Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs).
GR achieves a higher GPT-judged win-rate in RLHF, avoids overly focusing on the format in rule-based math rewards, and prevents hacking the judge in LLM-as-a-Judge math tasks.
