Kaiyi Guo

Kenny676's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted 4 papers about 15 hours ago

False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents

Paper • 2609.39102 • Published 6 days ago • 470

The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation

Paper • 2609.36484 • Published 7 days ago • 552

LoopVL: Recurrent Visual Intelligence

Paper • 2609.38426 • Published 7 days ago • 461

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

Paper • 2602.02474 • Published Feb 2 • 64

upvoted 2 papers 3 days ago

Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL

Paper • 2610.00574 • Published 6 days ago • 59

EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery

Paper • 2609.40340 • Published 6 days ago • 107

upvoted a collection 4 days ago

ReasonFLux-Coder

Collection

Coding LLMs excel at both writing code and generating unit tests. • 9 items • Updated 17 days ago • 12

upvoted 3 papers 6 days ago

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published 15 days ago • 222

HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?

Paper • 2609.01437 • Published Sep 1 • 567

The Verification Horizon: No Silver Bullet for Coding Agent Rewards

Paper • 2606.26300 • Published Jun 24 • 53

upvoted 2 papers 8 days ago

Training Object Permanence in World Models

Paper • 2609.28654 • Published 13 days ago • 243

StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback

Paper • 2402.01391 • Published Feb 2, 2024 • 44

upvoted an article 14 days ago

view article

Article

A Guide to Reinforcement Learning Post-Training for LLMs: PPO, DPO, GRPO, and Beyond

• 

karina-zadorozhny

•

Jan 19

• 52

upvoted a collection 19 days ago

SFTvsRL Models & Data

Collection

This collection contains 4 initial checkpoints for https://github.com/LeslieTrue/SFTvsRL and necessary data for V-IRL training. • 6 items • Updated Mar 2 • 10

upvoted a paper 26 days ago

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training

Paper • 2501.17161 • Published Jan 28, 2025 • 128