IACM-RL: Intent-Aware Context Management and Reinforcement Learning for Complex Tool Invocation under Dynamic Intent Fluctuations

arXiv:2608.02110 · cs.CL, cs.AI · Submitted 2026-08-03 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-08-03

Updated: 2026-10-14

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Reinforcement Learning — 50× in this paper · explained in 4 episodes
• Context Manager — 23× in this paper
• Benchmark — 17× in this paper · explained in 2 episodes
• Accuracy — 11× in this paper · explained in 1 episode
• Core — 11× in this paper · explained in 2 episodes
• Augmentation — 9× in this paper · explained in 2 episodes
• Generalization — 8× in this paper · explained in 14 episodes
• Robustness — 8× in this paper · explained in 6 episodes
• Consistency — 7× in this paper · explained in 3 episodes
• LLM judge — 6× in this paper · explained in 2 episodes
• Ablation — 5× in this paper · explained in 1 episode
• Distillation — 5× in this paper · explained in 2 episodes
• Editor — 5× in this paper · explained in 2 episodes
• Calibration — 4× in this paper · explained in 9 episodes
• Composition — 4× in this paper · explained in 2 episodes
• Credit assignment — 4× in this paper · explained in 3 episodes
• Actor — 3× in this paper · explained in 3 episodes
• Decode — 3× in this paper · explained in 2 episodes
• Graph Construction — 3× in this paper · explained in 3 episodes
• Hyperparameters — 3× in this paper · explained in 2 episodes
• Joint Goal Accuracy — 3× in this paper
• Trace — 3× in this paper · explained in 2 episodes
• Abstraction — 2× in this paper · explained in 1 episode
• Decoupling — 2× in this paper · explained in 2 episodes
• Dependency Tool Chain Rate — 2× in this paper
• Hallucination — 2× in this paper · explained in 4 episodes
• Intent Switch Success Rate — 2× in this paper
• LLM-as-a-Judge — 2× in this paper · explained in 4 episodes
• Papers — 2× in this paper · explained in 3 episodes
• Personalization — 2× in this paper · explained in 2 episodes
• Redundant and Infinite-loop Rate — 2× in this paper
• Stale Context Residual Rate — 2× in this paper
• State Management — 2× in this paper · explained in 3 episodes
• Two-stage training — 2× in this paper · explained in 2 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Ablation Study — 1× in this paper · explained in 3 episodes
• Active Clarification Reward — 1× in this paper
• Benchmarking — 1× in this paper · explained in 4 episodes
• Catastrophic Forgetting — 1× in this paper · explained in 9 episodes
• Evaluation Dimensions — 1× in this paper · explained in 3 episodes
• Evaluation Framework — 1× in this paper · explained in 3 episodes
• Evaluation Protocol — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Function Sequence Patterns — 1× in this paper
• Language Models (LLMs) — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 9 episodes
• Memorization — 1× in this paper · explained in 4 episodes
• Multi-Agent System — 1× in this paper · explained in 7 episodes
• Overfitting — 1× in this paper · explained in 2 episodes
• POMDP — 1× in this paper · explained in 2 episodes
• Perplexity — 1× in this paper · explained in 3 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 26 episodes
• Student Policy — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Technical Report: Evaluating Goal Drift in Language Model Agents
• $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
• Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction
• Intent Assurance using LLMs guided by Intent Drift
• ACON: Optimizing Context Compression for Long-horizon LLM Agents
• LLMs Get Lost In Multi-Turn Conversation
• Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents
• AgentAsk: Multi-Agent Systems Need to Ask
• AgentBench: Evaluating LLMs as Agents
• Probing the Lack of Stable Internal Beliefs in LLMs
• Same Signal, Different Semantics: A Cross-Framework Behavioral Analysis of Software Engineering Agents
• EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training
• Asymmetric Goal Drift in Coding Agents Under Value Conflict
• Evolutionary Context Search for Automated Skill Acquisition
• The OpenHands Software Agent SDK: A Composable and Extensible Foundation for Production Agents
• When Should Models Change Their Minds? Contextual Belief Management in Large Language Models
• SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
• WebArena: A Realistic Web Environment for Building Autonomous Agents
• MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
• DFPO: Scaling Value Modeling via Distributional Flow towards Robust and Generalizable LLM Post-Training

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering