arXiv Machine Learning

Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning

arXiv:2608. 00301v1 Announce Type: new Abstract: Error-penalized scoring rules ($+1$ for a correct answer, $-\lambda$ for a wrong one, $0$ for abstaining) are increasingly prescribed against hallucination: a rational agent facing such a rule answers exactly when its correctness probability exceeds Chow's threshold $t^\ast=\lambda/(1+\lambda)$.

arXiv Machine Learning

Aug 5

The Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and The Channel, Not the Content, Decides What Works

arXiv:2607. 21273v2 Announce Type: replace Abstract: Dense per-step supervision is the standard remedy for sparse-reward long-horizon LLM agents: reward the policy for predicting its next observation, which looks provably safe under potential-based shaping.

By Yu Wang

llms agents

arXiv AI

Jun 30

Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models

arXiv:2606. 30627v1 Announce Type: cross Abstract: Conservative offline training is widely advocated as a safe foundation for subsequent online adaptation: if a policy stays close to well-supported behaviour, the argument goes, it is less likely to exploit imperfections in a learned reward model.

By Subramanyam Sahoo, Aman Chadha, Vinija Jain, Divya Chaudhary

reinforcement-learning safety

arXiv AI

Jul 9

RLVP: Penalize the Path, Reward the Outcome

arXiv:2607. 07435v1 Announce Type: cross Abstract: Agents acting on our behalf in the real world (e.

By Bojie Li, Noah Shi

agents reinforcement-learning

arXiv Machine Learning

Jul 24

The Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and What Actually Works

arXiv:2607. 21273v1 Announce Type: new Abstract: Dense per-step supervision is an appealing remedy for sparse-reward, long-horizon LLM agents: reward the agent for predicting its next observation, and memory should follow.

By Yu Wang

llms agents reinforcement-learning benchmarks

arXiv AI

1d ago

When to Communicate: Belief Distributions and KL Divergence for Principled Gating in Multi-Agent RL

arXiv:2608. 14559v1 Announce Type: new Abstract: Effective communication in multi-agent reinforcement learning requires agents to decide not only \textit{what} to communicate, but when?

By Teoman Kaman

agents reinforcement-learning benchmarks

arXiv AI

Jul 7

Gradient Regularization Mitigates Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards

arXiv:2602. 18037v2 Announce Type: replace-cross Abstract: Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs).

By Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, Masashi Sugiyama

llms reinforcement-learning safety

arXiv AI

Jul 15

A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism

arXiv:2607. 12640v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards, and Group Relative Policy Optimization (GRPO) in particular, is now run routinely on a supervised checkpoint in the hope of producing a stronger agent.

By Chengguang Gan, Zhixi Cai, Yunhao Liang, Hanjun Wei, Shiwen Ni, Qinghao Zhang

llms rag agents reinforcement-learning multimodal

arXiv AI

Jun 3

When RLHF Fails: A Mechanistic Taxonomy of Reward Hacking, Collapse, and Evaluator Gaming

arXiv:2606. 03238v1 Announce Type: cross Abstract: Reinforcement learning from human feedback (RLHF) makes large-scale post-training possible by replacing an underspecified human objective with learned and scalable proxies.

By Zelalem Abahana

llms reinforcement-learning safety

arXiv Machine Learning

Jul 21

The Behavioral Credibility Trilemma: When Calibrated Autonomy Becomes Impossible

arXiv:2605. 25739v2 Announce Type: replace Abstract: We prove that no reinforcement learning policy with confidence-gated autonomy can simultaneously achieve maximum helpfulness, optimal calibration, and full autonomy under rational oversight, whenever some tasks exceed the agent's reliable competence: the Behavioral Credibility Trilemma.

By Lauri Lov\'en, Nam Do, Hassan Mehmood, Dinesh Kumar Sah, Sasu Tarkoma

agents reinforcement-learning

arXiv AI

Jul 7

Regime-Conditional Stabilisation of LLM-Augmented Cooperative Multi-Agent Reinforcement Learning

arXiv:2607. 04470v1 Announce Type: cross Abstract: Large Language Models (LLMs) offer a natural interface for translating human objectives into reward signals for cooperative multi-agent reinforcement learning (MARL), yet the training-time dynamics of this integration remain poorly understood.

By Faid Keddouri, Sohaib Houhou, Aissa Boulmerka, Nadir Farhi

llms agents reinforcement-learning

arXiv AI

Jul 21

Teach it to stop, not just to click

arXiv:2607. 17136v1 Announce Type: cross Abstract: Agentic computer-use RL is reported in single runs, and those numbers mislead.

By Barada Sahu (Cabal AI), Shivesh Pandey (Para AI)

agents reinforcement-learning efficiency multimodal

arXiv Machine Learning

Aug 11

Quality-Diversity Stress Tests for Process Reward Models:What Archive Coverage Can and Cannot Certify

arXiv:2608. 08008v1 Announce Type: new Abstract: Process reward models (PRMs) score intermediate reasoning steps and are widely used for search, ranking, and training, but optimization can exploit these learned proxies by increasing reward while turning correct reasoning into incorrect reasoning.

By Ibne Farabi Shihab, Fariya Afrin

fine-tuning safety

Related stories