State-Grounded Conditioning: Wrapping User-Facing LLM Agents Where Direction Depends on Live State
cs.AI
Submitted: 2026-09-23
Updated: 2026-09-23
Comments: 11 pages (6-page main body + Limitations, Ethics, References, Appendix); 4 figures; 3 tables. Preprint. Under review at EACL 2027 (Industry Track)
Project page: https://picrew.github.io/LLM-Harness
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding
- Agents Thinking Fast and Slow: A Talker-Reasoner Architecture
- Self-Refine: Iterative Refinement with Self-Feedback
- FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation
- A Survey on Large Language Model-Based Game Agents
- The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input
- MemGPT: Towards LLMs as Operating Systems
- Survey of Hallucination in Natural Language Generation
- A Large-Scale Real-World Evaluation of LLM-Based Virtual Teaching Assistant
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- SLOT: Structuring the Output of Large Language Models
- Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
- WebArena: A Realistic Web Environment for Building Autonomous Agents
- LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
- TravelPlanner: A Benchmark for Real-World Planning with Language Agents
- ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models
- Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents
- Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support
- Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
Related papers
- 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
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems