arXiv.org
arXiv.org is Agent harnesses repeatedly invoke reasoning after tool feedback, even when an earlier call has already fixed the next action. We introduce HarnessMM (Harness Minus Minus), a method for improving harness token…. It is ranked #61 on the Vibe Coding Radar, in Agent infra & costs, first seen 10 days ago and shared in 3 posts (121.6k views).
[2610.32813] HarnessMM: Recursive Self-Improvement for Harness Token Efficiency Skip to main content Search Search arXiv Press Enter to search · Advanced search --> Computer Science > Artificial Intelligence arXiv:2610.32813 (cs) [Submitted on 23 Oct 2026] Title: HarnessMM: Recursive Self-Improvement for Harness Token Efficiency Authors: Yuhua Jiang , Chang Xu , Feifei Gao , Li Zhao View a PDF of the paper titled HarnessMM: Recursive Self-Improvement for Harness Token Efficiency, by Yuhua Jiang and 3 other authors View PDF HTML (experimental) Abstract: Agent harnesses repeatedly invoke reasoning after tool feedback, even when an earlier…
What people said about arXiv.org on X
🫨简直无敌！DeepSeek Harness被清华微软团队直接“瘦身”：Token消耗砍掉87.93%，成功率反而暴涨14%！ Agent每次工具反馈后都要重新问一遍LLM，哪怕下一步早定死了（写完代码→编译，编译成功→测试）。清华+微软团队用HarnessMM直接把这些“已决定”的步骤缓存下来，跳过重复推理。 核心成果（98个HumanEval任务）： 🔹DeepSeek Harness：Token从96.28k降到11.62k，砍掉87.93%！成功率从71.43%直接飙到85.71% 🔹Codex：成功率稳在90.82%，Token再降29.54%（66.72k→47.02k）…
— @NFT_Chen, 4 days ago · 1k likes · see the post
Learn to build a custom harness, folks! There is so much alpha in understanding how to optimize the harness. This work shows the benefits of a smart harness configuration and caching the decisions your agent harness has already made. (bookmark it) They propose HarnessMM, which cuts tokens per task on a DeepSeek…
— @omarsar0, 10 days ago · 426 likes · see the post
很多人还对梁文峰不甚了解！一个视频带你回顾梁文峰的来时路！ 梁文峰：我不是“圣人”！我的理想有迹可循，我的事迹水到渠成！我的事业必将成功！ 人不以赚钱为目的去做事，就能赚到源源不断的钱！ #DeepSeek #梁文峰 #梁圣 #大胖鱼 https://x.com/NFT_Chen/status/2117182992389157027?a=cCRffLfbLS9VgagUH38&s=20
— @NFT_Chen, 4 days ago · 55 likes · see the post
Alternatives to arXiv.org
- Aether — It’s free to start.
- deepseek-harness — I have all three of my AI agent leads for Hermes, Grok Bot, and OpenClaw pouring over the repos for DeepSeeks proposed…
- Entire — Source control is at a breaking point. Today, we open the portal to a new kind of Git forge: reimagined review, CI…
- Neon — We wrote about the changes, our MCP usage numbers, and the spec update
- Agent37 — just made the hosting cheaper
- boat — Persistent Linux VMs for AI agents with SSH, Docker, and fast fork. Billed per second, free when stopped. 7-day free…
arXiv.org in numbers
- Rank on the Vibe Coding Radar: #61 of 2591
- Shared in 3 posts by 2 accounts: @NFT_Chen, @omarsar0
- 121.6k views on those posts
- First seen 10 days ago, last shared 4 days ago
- Pricing seen by AGK Intelligence: free
- Market: Agent infra & costs
FAQ
What is arXiv.org?
Agent harnesses repeatedly invoke reasoning after tool feedback, even when an earlier call has already fixed the next action. We introduce HarnessMM (Harness Minus Minus), a method for improving harness token… It was first shared on X 10 days ago and is ranked #61 on the Vibe Coding Radar.
Is arXiv.org free?
Yes, it is free to use.
Who shared arXiv.org?
2 accounts on X, including @NFT_Chen, @omarsar0, in 3 posts totalling 121.6k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 7.9k posts from 5k X accounts over the last 14 days, 2.6k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 13:22 UTC. Full method.