The papers

Most central to this idea first, not most recent.

• Big Techcs.LG

  Exactly-once behavior shifts from models to contracts with fault type

  Microsoft · Sept 2026

• Research Labcs.CR

  Local LLM agents can erase the traces meant to audit them

  ELLIS Institute Tübingen, Max Planck Institute for Intelligent Systems · Sept 2026

• Top Universitycs.AI

  Task-specific plans and terminal checks create different value for LLM agents

  The Chinese University of Hong Kong, University of Edinburgh · Sept 2026

• Industrycs.AI

  Coding-agent harness choices should match model skill and context budget

  UMass Amherst, Zoom Video Communications · Sept 2026

• Top Universitycs.CR

  Malicious plugin updates can hijack AI agent lifecycle hooks

  Beijing University of Posts and Telecommunications, Data Technology Support Center of the Cyberspace Administration of China · Sept 2026

• Big Techcs.AI

  Recursive harness search cuts coding-agent token use nearly in half

  NVIDIA, NTU · Sept 2026

• Independentcs.SE

  LLMs can build agent harnesses, but improvements transfer poorly

  Sept 2026

• Independentcs.RO

  A semantic harness lets vision-language models control different robots

  Sept 2026

• Industrycs.CL

  Controlled harness reveals what helps agents refactor entire repositories

  Pi School, Translated · Sept 2026

• Independentcs.AI

  Persistent evidence tracking improves language-agent performance on bioinformatics benchmarks

  Sept 2026

• Top Universitycs.CR

  Agent harnesses can elevate malicious context past instruction defenses

  Nanjing University, Honor Device Co., Ltd · Aug 2026

• Chinese Techcs.AIcode

  Models remain weak at steering coding agents through long tasks

  DreamX Team, Alibaba Group · Sept 2026

• Top Universitycs.AI

  Markdown interfaces cut agent context costs without reducing task success

  Seoul National University, H1R.AI · Sept 2026

• Chinese Techcs.SE

  Self-evolving harness turns general AI agents into stronger RCA specialists

  The Chinese University of Hong Kong, Individual Researcher · Aug 2026

• Chinese Techcs.AI

  Self-improving context programs help frozen models understand long videos

  Tencent · Aug 2026

• Top Universitycs.CL

  Generating Agent Harnesses on Demand Improves Models Across Benchmarks

  LV-NUS Lab · Aug 2026

• Big Techcs.AI

  New framework lets you train AI agents inside the same harness systems they use at inference

  Columbia University, Dartmouth College · July 2026

• Top Universitycs.LGcode

  New benchmark standardizes evaluation of coding agent harnesses

  TokenRhythm Technologies, Infinigence AI · June 2026

• Industrycs.AI

  Safe agent self-modification requires verifiable, expressive recovery mechanisms

  Independent Researcher · Sept 2026

• Top Universitycs.CR

  Agent safety requires persistent state across autonomous loop iterations

  University of Chinese Academy of Sciences, Nanyang Technological University · Aug 2026

• Industrycs.AI

  Layered agent harnesses improve software over repeated autonomous development cycles

  Shanghai Artificial Intelligence Laboratory · Sept 2026

• Industrycs.AI

  Distilled repository skills improve agents conducting machine-learning research

  Beijing Academy of Artificial Intelligence, University of Science and Technology of China · Sept 2026

• Top Universitycs.SE

  Embedded AI agents can adapt software during live use

  University of Kentucky, University College London · Sept 2026

• Top Universitycs.AI

  Behavior-aware checks make agent harness evolution cheaper and more reliable

  Fudan University, Shanghai Key Laboratory of Data Science · Aug 2026

• AI Startupcs.AIcode

  Goal pressure exposes wide gaps in security agents’ scope adherence

  dreadnode · Sept 2026

• Independentcs.CL

  Models struggle to improve themselves from vague capability goals

  Sept 2026