Patrick Jiang

pat-jj's activity

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

upvoted 2 papers 4 days ago

Learning from Teacher Continuations at Student States

Paper • 2609.36246 • Published 7 days ago • 40

Selecting Diverse SFT Traces Improves Post-RL Generalization

Paper • 2609.33780 • Published 8 days ago • 38

upvoted a paper 12 days ago

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published 14 days ago • 222

upvoted a paper about 1 month ago

Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning

Paper • 2609.03430 • Published Sep 3 • 188

upvoted 2 papers 4 months ago

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses

Paper • 2606.02373 • Published Jun 1 • 59

Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism

Paper • 2606.00408 • Published May 29 • 64

upvoted 2 papers 5 months ago

Learning to Predict Future-Aligned Research Proposals with Language Models

Paper • 2603.27146 • Published Apr 6 • 6

Retrieval is Cheap, Show Me the Code: Executable Multi-Hop Reasoning for Retrieval-Augmented Generation

Paper • 2605.12975 • Published May 13 • 9

upvoted a paper 8 months ago

Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning

Paper • 2602.01058 • Published Feb 1 • 45

upvoted a paper 10 months ago

Adaptation of Agentic AI

Paper • 2512.16301 • Published Dec 18, 2025 • 111

upvoted a paper 12 months ago

GRACE: Generative Representation Learning via Contrastive Policy Optimization

Paper • 2510.04506 • Published Oct 6, 2025 • 12

upvoted 3 papers over 1 year ago

s3: You Don't Need That Much Data to Train a Search Agent via RL

Paper • 2505.14146 • Published May 20, 2025 • 20

DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learning

Paper • 2503.00223 • Published Feb 28, 2025 • 2

RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation

Paper • 2502.10996 • Published Feb 16, 2025 • 1