Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models face the "knowledge cutoff" problem because their parametric memory remains frozen after pretraining, preventing them from natively internalizing new information or tools on the fly. |
| Approach: | They propose a framework that supports modular skill transfer for efficient and effective knowledge adaptation by extracting a domain-agnostic **Skill Vector from a source domain. |
| Outcome: | Experiments on knowledge-incorporation QA (SQuAD, LooGLE) and agentic tool-use benchmarks show that the proposed framework outperforms state-of-the-art self-editing SFT by 9.9 points. |
Similar Papers
Reinforcement Learning for Self-Improving Agent with Skill Library (2026.acl-long)
Copied to clipboard
Jiongxiao Wang, Qiaojing Yan, Yawei Wang, Yijun Tian, Soumya Smruti Mishra, Zhichao Xu, Megha Gandhi, Panpan Xu, Lin Lee Cheong
| Challenge: | Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in complex reasoning and multi-turn interactions but struggle to continuously improve and adapt when deployed in new environments. |
| Approach: | They propose a Reinforcement Learning-based approach to enhance agents’ self-improvement capabilities with a skill library. |
| Outcome: | The proposed framework achieves 8.9% higher Scenario Goal Completion when applied to supervised-finetuned model with expert experience while requiring 26% fewer interaction steps and generating 59% fewer tokens. |
Data Doping or True Intelligence? Evaluating the Transferability of Injected Knowledge in LLMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a study shows that comprehension-intensive fine-tuning tasks retain knowledge longer . however, all models exhibit significant performance drops when applying injected knowledge in broader contexts . |
| Approach: | study: comprehension-intensive fine-tuning tasks achieve higher knowledge retention rates . larger models show improved retention across all task types, study finds . |
| Outcome: | a new study shows that comprehension-intensive fine-tuning tasks retain knowledge better than mapping-oriented tasks despite exposure to identical factual content. |
Tool Zero: Training Tool-Augmented LLMs via Pure RL from Scratch (2025.findings-emnlp)
Copied to clipboard
Yirong Zeng, Xiao Ding, Yutai Hou, Yuxian Wang, Li Du, Juyi Dai, Qiuyang Ding, Duyu Tang, Dandan Tu, Weiwen Liu, Bing Qin, Ting Liu
| Challenge: | Experimental results demonstrate that our models achieve over 7% performance improvement compared to both SFT and RL-with-SFT models under the same experimental settings. |
| Approach: | They propose a dynamic generalization-guided reward design for rule-based RL that shifts rewards from exploratory to exploitative tool-use patterns. |
| Outcome: | The proposed model achieves over 7% performance improvement compared to SFT and RL-with-SFT models under the same experimental settings. |
Investigating Cross-Modal Skill Injection: Scenarios, Methods, and Hyperparameters (2026.acl-long)
Copied to clipboard
| Challenge: | Existing research lacks systematic analysis of the applicability and methodology of cross-modal skill injection. |
| Approach: | They investigate the applicability and methodology of cross-modal skill injection by integrating a domain-expert LLM into a VLM. |
| Outcome: | The proposed method enables transfer of domain-specific expertise from Large Language Models (LLMs) to VLMs without incurring additional training data requirements or significant computational overhead. |
From English to Second Language Mastery: Enhancing LLMs with Cross-Lingual Continued Instruction Tuning (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) acquire strong language skills through extensive pre-training and supervised fine-tuning (SFT) on instructionresponse pairs. |
| Approach: | They propose a method which leverages translation-based parallel instruction data to enhance cross-lingual adaptability. |
| Outcome: | The proposed model improves on Llama-2-7B across five languages against three objective benchmarks and an LLM-as-a-judge benchmark. |
Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches allocate an equal number of rollouts to all questions during the RL process, which is inefficient. |
| Approach: | They propose a mechanism for dynamically allocating rollout budgets based on the difficulty of the problems, enabling more efficient RL training. |
| Outcome: | The proposed model improves response precision while preserving exploratory ability to uncover potential correct pathways. |
Rethinking Expert Trajectory Utilization in LLM Post-training for Mathematical Reasoning (2026.acl-long)
Copied to clipboard
Bowen Ding, Yuhan Chen, Jiayang Lyu, Jiyao Yuan, Qi Zhu, Shuangshuang Tian, Dantong Zhu, Futing Wang, Heyuan Deng, Fei Mi, Lifeng Shang, Tao Lin
| Challenge: | evolving generic Large Language Models into specialized Large Reasoning Models requires effective post-training. |
| Approach: | They propose a plasticity-ceiling framework to harness expert trajectories . they establish the Sequential SFT-then-RL pipeline as the superior standard . |
| Outcome: | The proposed framework overcomes stability and premature convergence deficits in synchronized approaches. |
Semi-supervised Fine-tuning for Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing LLMs require labeled data, which can be costly in real-world applications. |
| Approach: | They propose a framework that can fully exploit labeled and unlabeled data for LLM fine-tuning . they conducted experiments using GPT-4o-mini and Llama-3.1 on seven general or domain-specific datasets . |
| Outcome: | The proposed framework can fully exploit labeled and unlabeled data for LLM alignment from a propagate-and-select manner. |
Behavior Knowledge Merge in Reinforced Agentic Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for supervised fine-tuning (SFT) are suboptimal to preserve task-specific capabilities on RL-trained agentic models. |
| Approach: | They propose a distribution-aware merging framework specifically designed for RL-trained agentic models that disentangles shared and task-specific unique parameter updates while selectively preserving and rescaling unique ones. |
| Outcome: | Experiments across multiple agent domains and model architectures show that the proposed framework surpasses baselines and unlocks synergistic potential among agents. |
Representation Interventions Enable Lifelong Knowledge Memory Control in LLMs (2026.acl-long)
Copied to clipboard
Xuyuan Liu, Shengyu Chen, Xinshuai Dong, Yanchi Liu, Xujiang Zhao, Haoyu Wang, Yujun Yan, Haifeng Chen, Zhengzhang Chen
| Challenge: | Large language models (LLMs) produce outdated or inaccurate content. Updating their knowledge efficiently and accurately without costly retraining is a major challenge. |
| Approach: | They propose a robust and scalable method that treats knowledge control as interventions within the model’s representation space. |
| Outcome: | The proposed method achieves fine-grained control over complex, unstructured knowledge while maintaining general utility with frozen base weights. |