Image de couverture de livre

MAIGO: Mitigating Lost-in-Conversation with History-Cleaned On-Policy Self-Distillation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zheng, Haoyu, Zhu, Yun, Yuan, Shu, Chen, Shangming, Wang, Qing, Zhang, Wenqiao, Xiao, Jun, Zhuang, Yueting
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

PILOT: Planning via Internalized Latent Optimization Trajectories for Large Language Models
par: Zheng, Haoyu, et autres
Publié: (2026)

Bridging Local Details and Global Context in Text-Attributed Graphs
par: Wang, Yaoke, et autres
Publié: (2024)

Fast Thinking for Large Language Models
par: Zheng, Haoyu, et autres
Publié: (2025)

Enhancing Post-Training Quantization via Future Activation Awareness
par: Lv, Zheqi, et autres
Publié: (2026)

IDEAL: Leveraging Infinite and Dynamic Characterizations of Large Language Models for Query-focused Summarization
par: Cao, Jie, et autres
Publié: (2024)

DuetRAG: Collaborative Retrieval-Augmented Generation
par: Jiao, Dian, et autres
Publié: (2024)

SimCT: Recovering Lost Supervision for Cross-Tokenizer On-Policy Distillation
par: Sun, Jie, et autres
Publié: (2026)

Self-Distilled Agentic Reinforcement Learning
par: Lu, Zhengxi, et autres
Publié: (2026)

Meta-Reflection: A Feedback-Free Reflection Learning Framework
par: Wang, Yaoke, et autres
Publié: (2024)

EchoDistill:Alignment Noisy-to-Clean Self-Distillation for Robust Audio LLMs
par: Lin, Liang, et autres
Publié: (2026)

Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs
par: Dai, Yang, et autres
Publié: (2025)

Mitigating Lost in Multi-turn Conversation via Curriculum RL with Verifiable Accuracy and Abstention Rewards
par: Li, Ming, et autres
Publié: (2025)

MoA: Heterogeneous Mixture of Adapters for Parameter-Efficient Fine-Tuning of Large Language Models
par: Cao, Jie, et autres
Publié: (2025)

History-Aware Conversational Dense Retrieval
par: Mo, Fengran, et autres
Publié: (2024)

Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives
par: Zhang, Wenqi, et autres
Publié: (2024)

Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models
par: Zhao, Siyan, et autres
Publié: (2026)

Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation
par: Zhang, Junhao, et autres
Publié: (2025)

Milestone-Guided Policy Learning for Long-Horizon Language Agents
par: Wang, Zixuan, et autres
Publié: (2026)

Hierarchical Budget Policy Optimization for Adaptive Reasoning
par: Lyu, Shangke, et autres
Publié: (2025)

MT-OSC: Path for LLMs that Get Lost in Multi-Turn Conversation
par: Singh, Jyotika, et autres
Publié: (2026)

Mitigating Lost-in-Retrieval Problems in Retrieval Augmented Multi-Hop Question Answering
par: Zhu, Rongzhi, et autres
Publié: (2025)

Think Twice, Click Once: Enhancing GUI Grounding via Fast and Slow Systems
par: Tang, Fei, et autres
Publié: (2025)

LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimization
par: Wu, Xingyu, et autres
Publié: (2025)

Let LRMs Break Free from Overthinking via Self-Braking Tuning
par: Zhao, Haoran, et autres
Publié: (2025)

Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning
par: Zhao, Zhengyang, et autres
Publié: (2026)

Intent Mismatch Causes LLMs to Get Lost in Multi-Turn Conversation
par: Liu, Geng, et autres
Publié: (2026)

Crosslingual On-Policy Self-Distillation for Multilingual Reasoning
par: Liu, Yihong, et autres
Publié: (2026)

PSC: Extending Context Window of Large Language Models via Phase Shift Calibration
par: Zhu, Wenqiao, et autres
Publié: (2025)

Detection and Mitigation of Hallucination in Large Reasoning Models: A Mechanistic Perspective
par: Sun, Zhongxiang, et autres
Publié: (2025)

CrossView Suite: Harnessing Cross-view Spatial Intelligence of MLLMs with Dataset, Model and Benchmark
par: Wang, Wei, et autres
Publié: (2026)

LLMs Get Lost In Multi-Turn Conversation
par: Laban, Philippe, et autres
Publié: (2025)

Cognitive Policy-Driven LLM for Diagnosis and Intervention of Cognitive Distortions in Emotional Support Conversation
par: Zhong, Lin, et autres
Publié: (2026)

Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning
par: Wu, Tong, et autres
Publié: (2025)

Are Full Rollouts Necessary for On-Policy Distillation?
par: Zhang, Yaocheng, et autres
Publié: (2026)

Towards Distillation-Resistant Large Language Models: An Information-Theoretic Perspective
par: Fang, Hao, et autres
Publié: (2026)

OPSDL: On-Policy Self-Distillation for Long-Context Language Models
par: Zhang, Xinsen, et autres
Publié: (2026)

Chart-HQA: A Benchmark for Hypothetical Question Answering in Charts
par: Chen, Xiangnan, et autres
Publié: (2025)

VisualThink-VLA: Visual Intermediate Reasoning for Effective and Low-Latency Vision-Language-Action Policies
par: Gao, Mingjian, et autres
Publié: (2026)

MathClean: A Benchmark for Synthetic Mathematical Data Cleaning
par: Liang, Hao, et autres
Publié: (2025)

Internalize the Temperature: On-Policy Self-Distillation as Policy Reheater for Reinforcement Learning
par: Yang, Xuewei, et autres
Publié: (2026)