Been heads-down lately, but I finally had a couple of days to catch up on some recent papers. Sharing...
Been heads-down lately, but I finally had a couple of days to catch up on some recent papers. Sharing one of them here.
Milestone-Guided Policy Learning for Long-Horizon Language Agents
Zhejiang University (ZJU-REAL) · arXiv:2605.06078 · github.com/ZJU-REAL/BEACON
It tackles a very concrete and fairly painful problem: when you train an agent on long tasks with RL, it doesn't just get a bit worse — it falls off a cliff as the horizon grows.
TL;DR
I've been reading BEACON from Zhejiang University's ZJU-REAL lab (arXiv:2605.06078), and the more I...
RL‑Driven Agentic Optimization Training agents with only sparse rewards often yields unstable...
A Blog post by LinkedIn on Hugging Face
Why trajectory-level reward is a terrible teacher for multi-step agents - and how a 2026 paper called SDAR proposes to fix it.…
I have a small cooperative multi-agent reinforcement-learning setup: eight agents on a 100×100 grid,...
Your broken harness is actively making the model worse. Here's what I keep seeing after years of eyeballing trajectories, and…
