Connect the Dots: Training LLMs for Long-Lifecycle Agents with Cross-Domain Generalization Via Reinforcement Learning
cs.LG, cs.AI, cs.CL
Submitted: 2026-06-18
Updated: 2026-09-20
Comments: arXiv v2 updates: scale up experiments to larger Qwen3.6/3.8 models and new domains; support multi-teacher on-policy distillation; add theoretical study of learning dynamics
Code: https://github.com/agentscope-ai/Trinity-RFT
Project page: https://lifelongagent.github.io/iclr26.html
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Kimi k1.5: Scaling Reinforcement Learning with LLMs
- OpenAI o1 System Card
- Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks