ReHarness paper on arXiv: recurrent harness for agentic continual learning
rohanpaul_ai · x · 2026-11-13
The paper "Recurrent Harness for Agentic Continual Learning" (arXiv:2610.12668) comes from MIT, Meta MRS Research, and Yale.
Existing agent harnesses either evolve state only through forward information flow or need extra replay/validation data. ReHarness adds backward revision under a strictly online streaming protocol: after each instance, the resulting trajectory is used as feedback to revise the reflections retrieved for that instance, propagating feedback backward through the textual harness state.
Across three continual learning benchmarks and three model backbones, ReHarness consistently beats raw-context baselines, memory/reflection methods, and prompt optimization, producing source-grounded revisions that correct overgeneralized or outdated experience while improving stability and plasticity.
Related event: MIT and Meta Propose ReHarness for Agentic Continual Learning(2 posts)→
More from Research
- One Chart Explains the Linear Algebra Behind Modern AI, from Vectors to SVD — HankYeomans · 2026-11-13
- Codex agent shrinks a 27M-param transformer to 11.46M with equal loss via weight sharing — yoshiyama_akira · 2026-11-13
- Google open-sources Harness²: Agent = Model + Harness, recursive agent harnessing framework — 2C_ornot2C · 2026-11-13
- Question's Gambit: first-step reranking lifts agentic search accuracy to 90.5% — CShorten30 · 2026-11-13
- Terence Tao: major Riemann hypothesis progress came from AI-generated ideas nobody yet understands — burny_tech · 2026-11-13
- Deleting under 0.05% of weights matches full fine-tuning, NYU paper shows — burny_tech · 2026-11-13