Termination Lag: Coding Agents Solve Early but Stop Late
cs.AI, cs.CL
Submitted: 2026-11-11
Updated: 2026-11-11
Terminology
Sources
- Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
- SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
- More with Less: An Empirical Study of Turn-Control Strategies for Efficient Coding Agents
- Monitor Accuracy Isn't the Answer: A Decision-Value Framework for Evaluating AI Agent Restart Monitors
- Coherence Collapse: Diagnosing Why Code Agents Fail After Reaching the Right Code
- TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents
- Understanding Code Agent Behaviour: An Empirical Study of Success and Failure Trajectories
- Teaching Agents to Code Reliably
- AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation
- Executable Prefix Values for Locating Agent Failures
- EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction
- Babbling Suppression: Making LLMs Greener One Token at a Time
- Towards Green AI: Decoding the Energy of LLM Inference in Software Development
- Continue or Restart? The Limits of Early Failure Prediction in Coding Agents
- Fail-Fast, Restart-Smart: Early Failure Prediction and Restart for SWE Agentic Tasks
- Prompt-Induced Waste in Coding Agents: Reasoning, Effort, Harness Design, and End-to-End Cost
- The Best Programming Language for Tokenmaxxing: An Investigation of Coding Agent Behavior Across Programming Languages
- Agentless: Demystifying LLM-based Software Engineering Agents
- Failure as a Process: An Anatomy of CLI Coding Agent Trajectories
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents