arXiv:2606. 05263v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards improves reasoning and tool use, yet long-horizon language agents still learn unsupported evidence chains, belief drift, and shortcut actions that satisfy terminal checks.
RGDT-Bench is a new benchmark that evaluates large language models on Rule‑Governed Decision Tasks, where models must apply external rules to facts, justify decisions, and provide checkable justifications. The benchmark offers 202.1K condition‑level supervision slots across four task tracks and eight task‑probe combinations, and it labels warrant completeness through label‑blind extraction and deterministic checks. Evaluation shows that among correct responses, 40.2% of warrants are incomplete, and existing evaluators struggle to detect this, prompting the authors to train a reward model that improves AUROC to 69.24% and outperforms outcome‑supervised baselines.
The paper introduces a framework for evaluating AI systems that not only checks final labels but also tracks the reasoning behind them through three core sources—grounds, norms, and authority—forming an eight-cell counterfactual judgment cube. It defines minimal source replacement sets, called judgment receipts, to explain changes in verdicts and provides certification cost bounds for black-box evaluators. The authors present ReasonBench, a benchmark with 19,520 cases, and demonstrate that while high standard accuracy can mask robustness issues, receipt accuracy reveals significant gaps in reasoning consistency across different models.
The paper introduces Fact-Ablated Evaluation (FAE), a framework that iteratively removes cited evidence to test whether large language models (LLMs) adjust their fact‑checking predictions accordingly. Experiments reveal that many off‑the‑shelf LLMs rely more on internal knowledge than on the provided evidence. To address this, the authors propose REAL, a training method that uses counterfactual evidence supervision to encourage LLMs to base veracity judgments on evidence, achieving better evidence‑dependent performance across four datasets.
arXiv:2510. 19698v3 Announce Type: replace Abstract: Large Language Models (LLMs) can propose rules in natural language, sidestepping the need for a predefined predicate space in traditional rule learning.
Causal diagnostic models must explain how conclusions follow from evidence because diagnoses guide repairs and treatments. Yet serious cases are scarce, records rarely contain reasoning paths, and data transfer poorly across configurations, complicating local deployment.
arXiv:2608. 03674v1 Announce Type: new Abstract: Causal diagnostic models must explain how conclusions follow from evidence because diagnoses guide repairs and treatments.
arXiv:2608. 05235v1 Announce Type: cross Abstract: Research agents increasingly conduct multi-round machine-learning experiments in industrial recommendation settings and retain the resulting trajectories to guide later decisions.
arXiv:2607. 21806v1 Announce Type: new Abstract: Predictive machine learning (ML) models are increasingly used to aid human decision-makers across various high-risk domains such as healthcare and criminal justice.
The paper introduces Evidence‑Diagnosed Intervention Training (EDIT), a two‑phase framework designed to improve rubric‑faithful grading by large language models. EDIT‑SFT first identifies problematic reasoning steps using internal model signals—posterior belief over the final mark and input‑grounding scores—and revises only those steps with rubric checklists. EDIT‑RL then calibrates the grader with belief‑guided reward shaping, penalising harmful belief drifts while encouraging useful exploration. Experiments on two real‑world, multi‑subject grading benchmarks show that EDIT consistently outperforms strong supervised fine‑tuning and reinforcement learning baselines, with ablation studies confirming the importance of internal‑state diagnostics.
The paper introduces Boundary‑Calibrated Intervention Transfer (BCIT), a method for conditional experience transfer in autonomous large language model (LLM) post‑training. BCIT links each past update to its specific parent model, data, and training stage, checks whether those conditions still hold, vetoes updates with hard conflicts, and, when necessary, runs a bounded training trial to confirm applicability before adopting the update. Experiments on a 4B model across finance reasoning, text‑to‑SQL, and function calling show that BCIT reduces harmful updates and achieves higher final‑model quality under equal computational budgets compared to other approaches.
arXiv:2508.15754v2 Announce Type: replace-cross Abstract: Tool use is often assumed to monotonically improve reasoning, where external evidence is expected to help when relevant and be ignored when i...