arXiv Machine Learning June 2, 2026 By Lei Yang, Siyu Ding, Deyi Xiong

A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL

Read the original on arXiv Machine Learning →

arXiv:2606. 02398v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training improves large language models (LLMs) on individual domains such as mathematical reasoning, code generation, question answering, and creative writing (CW), but training on one domain often degrades performance on others.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Sep 10

One Step, One Lead: Mitigating Higher-Order Interference in Multi-Domain Reinforcement Learning via Cross-Step Control

The paper introduces OSOL, a method for mitigating higher‑order interference in multi‑domain reinforcement learning. OSOL selects a focus domain each iteration, uses token‑level footprints from the previous checkpoint to rank rebound risk, and applies an adaptively scaled correction to the GRPO update. Experiments on Qwen3‑30B‑A3B show a 5.7% improvement over the best baseline without higher‑order differentiation.

By Zihan Lin, Xiaohan Wang, Jie Cao, Jiajun Chai, Guojun Yin, Wei Lin, Ran He

llms reinforcement-learning safety

arXiv Machine Learning

Aug 24

Nothing Changed but the Model: CellFill -- Bounded In-Cell Learning for Bit-Identical, Revocable Updates to Quantized LLMs

arXiv:2608.20873v1 Announce Type: new Abstract: Every way of teaching a deployed language model something new -- full fine-tuning, adapter merging, model editing -- replaces the released checkpoint,...

By Zifeng Liu, Zhiyong Du, Yaxin Lu, Yiming Mao, Zhenhe Wang, Wenqi Shi, Zhengkun Jing

llms fine-tuning efficiency

arXiv AI

Jun 9

Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT?

arXiv:2605. 28860v2 Announce Type: replace-cross Abstract: Fine-tuning large language models (LLMs) frequently induces catastrophic forgetting of prior capabilities.

By Jeanmely Rojas Nunez, Viraj Sawant, Nathan Allen, Nomgondalai Amgalanbaatar, Yannis Zongo, Vasu Sharma, Maheep Chaudhary

llms reinforcement-learning fine-tuning

arXiv Machine Learning

Jul 2

Is One Layer Enough? Training A Single Transformer Layer Can Match Full-Parameter RL Training

arXiv:2607. 01232v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a central component of post-training large language models (LLMs), yet little is understood about how RL adaptation is distributed across transformer layers.

By Zijian Zhang, Rizhen Hu, Athanasios Glentis, Dawei Li, Chung-Yiu Yau, Hongzhou Lin, Mingyi Hong

llms agents reinforcement-learning

arXiv Machine Learning

Jun 16

Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning

arXiv:2601. 18699v2 Announce Type: replace Abstract: Sequential fine-tuning of Large Language Models (LLMs) adaptation to target tasks often triggers catastrophic forgetting, where the acquisition of novel target skills degrades ancestral capabilities.

By Gustav Olaf Yunus Laitinen-Fredriksson Lundstrom-Imanov

llms fine-tuning benchmarks safety

arXiv AI

Jun 9

Sparrow: Sparse Rollout for Stable and Efficient Long-context RL of Large Language Models

arXiv:2606. 08446v1 Announce Type: cross Abstract: Despite being powerful, reinforcement learning with verifiable rewards (RLVR) induces extremely long COT, making it computationally expensive.

By Yang Zhou, Ranajoy Sadhukhan, Zhaofeng Sun, Zhuoming Chen, Souvik Kundu, Saket Dingliwal, Sai Muralidhar Jayanthi, Aram Galstyan, Haizhong Zheng, Beidi Chen

llms reinforcement-learning fine-tuning efficiency

One email a morning, machine-written

Related stories

One Step, One Lead: Mitigating Higher-Order Interference in Multi-Domain Reinforcement Learning via Cross-Step Control

Nothing Changed but the Model: CellFill -- Bounded In-Cell Learning for Bit-Identical, Revocable Updates to Quantized LLMs

Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT?

Is One Layer Enough? Training A Single Transformer Layer Can Match Full-Parameter RL Training

Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning

Sparrow: Sparse Rollout for Stable and Efficient Long-context RL of Large Language Models