The papers

Most central to this idea first, not most recent.

• Big Techcs.AI

  Replayable news timelines test how LLM agents update forecasts

  Georgia Institute of Technology, Amazon · Sept 2026

• Industrycs.CRcode

  Cyber agents hide critical weaknesses behind successful attack workflows

  National Research Council, University of Windsor · Sept 2026

• Independentcs.AI

  Competitive games provide a dynamic benchmark for language model strategy

  Sept 2026

• Top Universitycs.AI

  Cross-device workflows expose major weaknesses in today’s GUI agents

  Beihang University, Beijing Institute of Technology · Sept 2026

• Big Techcs.AIcode

  Civilization benchmark exposes agents’ weak monitoring and long-term follow-through

  University of Oxford, Tony Blair Institute for Global Change · Sept 2026

• Top Universitycs.CVcode

  MLLMs fail to sustain goal-directed navigation in real-scale 3D city

  Shanghai Jiao Tong University, National University of Singapore · Aug 2026

• Industrycs.LG

  Domain context improves LLM agents’ time-series root-cause attribution

  ETH Zürich, CSEM SA · Aug 2026

• Independentcs.AI

  Benchmark tests whether coding agents improve governance without breaking repositories

  Sept 2026

• AI Startupcs.AIcode

  Goal pressure exposes wide gaps in security agents’ scope adherence

  dreadnode · Sept 2026

• Independentcs.SE

  Refresh schedules, not cache age, determine agent answer staleness

  Sept 2026

• Industrycs.IR

  Benchmark tests retrieval for agent-written queries across 190 million web pages

  Perplexity AI · Sept 2026

• Independentcs.SE

  LLMs can build agent harnesses, but improvements transfer poorly

  Sept 2026

• Industrycs.CL

  Tool-call traces expose extraction failures that source fidelity misses

  Infineon Technologies AG · Sept 2026

• Chinese Techcs.AIcode

  Models remain weak at steering coding agents through long tasks

  DreamX Team, Alibaba Group · Sept 2026

• Top Universitycs.CL

  LLM agents struggle to infer wellbeing from long-term wearable data

  Dartmouth College, University of Cambridge · Aug 2026

• Chinese Techcs.AI

  Process-level tests expose distinct mathematical agent capabilities in LLMs

  Sun Yat-sen University, Tencent Youtu Lab · Aug 2026

• Independentcs.SE

  Enterprise agents struggle when facts are hidden across business systems

  Sept 2026

• Top Universitycs.LGcode

  New benchmark standardizes evaluation of coding agent harnesses

  TokenRhythm Technologies, Infinigence AI · June 2026

• Independentcs.AI

  Deterministic guardrails expose gains that fool LLM-based judges

  Sept 2026

• Independentcs.CL

  LLM agents do not reliably improve from their own experience

  Sept 2026

• Independentcs.SEcode

  Coding agent benchmarks miss review compliance gaps

  Sept 2026

• Top Universitycs.CL

  Simulated user personas improve agents’ handling of multi-turn workflows

  State Key Laboratory of Multimedia Information Processing, Peking University · Sept 2026

• Big Techcs.AI

  Structured behavioral abstractions improve diagnosis of failures in LLM agents

  Tsinghua University, Microsoft Research · Sept 2026

• Industrycs.AI

  Dual-layer knowledge graph connects fragmented pharmaceutical process-development documents

  Sanofi US · Sept 2026

• Top Universitycs.AI

  New benchmark reveals coordination as distinct bottleneck for LLM agents

  University of Edinburgh, University of Oxford · June 2026