Image de couverture de livre

What and When to Distill: Selective Hindsight Distillation for Multi-Turn Agents

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Xiaozhe, Lyu, Tianyi, Li, Yang, Ma, Yichuan, Li, Peiji, Li, Linyang, Guo, Qipeng, Lin, Dahua, Chen, Kai
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic
par: Ma, Yichuan, et autres
Publié: (2026)

Beyond Mode Collapse: Distribution Matching for Diverse Reasoning
par: Li, Xiaozhe, et autres
Publié: (2026)

Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of Go
par: Ma, Yichuan, et autres
Publié: (2026)

Hindsight Hint Distillation: Scaffolded Reasoning for SWE Agents from CoT-free Answers
par: Wang, Shengjie, et autres
Publié: (2026)

ZipRL: Adaptive Multi-Turn Context Compression with Hindsight Response Replay
par: Hu, Zhexin, et autres
Publié: (2026)

From Multi-Agent to Single-Agent: When Is Skill Distillation Beneficial?
par: Xu, Binyan, et autres
Publié: (2026)

Turn Waste into Worth: Rectifying Top-$k$ Router of MoE
par: Zeng, Zhiyuan, et autres
Publié: (2024)

HEAL: Hindsight Entropy-Assisted Learning for Reasoning Distillation
par: Zhang, Wenjing, et autres
Publié: (2026)

MEND: Meta dEmonstratioN Distillation for Efficient and Effective In-Context Learning
par: Li, Yichuan, et autres
Publié: (2024)

Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs
par: Li, Xiaozhe, et autres
Publié: (2026)

HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents
par: Yeo, Woongyeng, et autres
Publié: (2026)

NP-Engine: Empowering Optimization Reasoning in Large Language Models with Verifiable Synthetic NP Problems
par: Li, Xiaozhe, et autres
Publié: (2025)

UnitCoder: Scalable Iterative Code Synthesis with Unit Test Guidance
par: Ma, Yichuan, et autres
Publié: (2025)

SD-Search: On-Policy Hindsight Self-Distillation for Search-Augmented Reasoning
par: Ma, Yufei, et autres
Publié: (2026)

Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections
par: Wang, Bo, et autres
Publié: (2025)

TL-GRPO: Turn-Level RL for Reasoning-Guided Iterative Optimization
par: Li, Peiji, et autres
Publié: (2026)

What Deserves Memory: Adaptive Memory Distillation for LLM Agents
par: Ma, Wenquan, et autres
Publié: (2025)

Escaping the Context Bottleneck: Active Context Curation for LLM Agents via Reinforcement Learning
par: Li, Xiaozhe, et autres
Publié: (2026)

FastMCTS: A Simple Sampling Strategy for Data Synthesis
par: Li, Peiji, et autres
Publié: (2025)

Bidirectional Distillation: A Mixed-Play Framework for Multi-Agent Generalizable Behaviors
par: Feng, Lang, et autres
Publié: (2025)

Hindsight Credit Assignment for Long-Horizon LLM Agents
par: Tan, Hui-Ze, et autres
Publié: (2026)

What are the Essential Factors in Crafting Effective Long Context Multi-Hop Instruction Datasets? Insights and Best Practices
par: Chen, Zhi, et autres
Publié: (2024)

Same Evidence, Different Answers: Canonical-Context On-Policy Distillation for Multi-Turn Language Models
par: Lin, Zizhuo, et autres
Publié: (2026)

Hindsight-Anchored Policy Optimization: Turning Failure into Feedback in Sparse Reward Settings
par: Wu, Yuning, et autres
Publié: (2026)

InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling
par: Li, Peiji, et autres
Publié: (2025)

AgentDistill: Training-Free Agent Distillation with Generalizable MCP Boxes
par: Qiu, Jiahao, et autres
Publié: (2025)

AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent
par: Luo, Yinyi, et autres
Publié: (2026)

Grounding and Enhancing Informativeness and Utility in Dataset Distillation
par: Wang, Shaobo, et autres
Publié: (2026)

Balanced Data Sampling for Language Model Training with Clustering
par: Shao, Yunfan, et autres
Publié: (2024)

Dark Distillation: Backdooring Distilled Datasets without Accessing Raw Data
par: Yang, Ziyuan, et autres
Publié: (2025)

When Single-Agent with Skills Replace Multi-Agent Systems and When They Fail
par: Li, Xiaoxiao
Publié: (2026)

TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agents
par: Wang, Jiaqi, et autres
Publié: (2026)

PANDO: Efficient Multimodal AI Agents via Online Skill Distillation
par: Li, Yubo, et autres
Publié: (2026)

GAOKAO-Eval: Does high scores truly reflect strong capabilities in LLMs?
par: Lei, Zhikai, et autres
Publié: (2024)

Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL
par: Li, Weizhen, et autres
Publié: (2025)

Prioritize Alignment in Dataset Distillation
par: Li, Zekai, et autres
Publié: (2024)

Foreground-Aware Dataset Distillation via Dynamic Patch Selection
par: Li, Longzhen, et autres
Publié: (2026)

Think, Speak, Decide: Language-Augmented Multi-Agent Reinforcement Learning for Economic Decision-Making
par: Ma, Heyang, et autres
Publié: (2025)

Efficient Personalized Reranking with Semi-Autoregressive Generation and Online Knowledge Distillation
par: Cheng, Kai, et autres
Publié: (2026)

H$^2$R: Hierarchical Hindsight Reflection for Multi-Task LLM Agents
par: Ye, Shicheng, et autres
Publié: (2025)