SELD: Detecting Silent Evidence Loss in Context Compression beyond the Answer
cs.CL, cs.AI
Submitted: 2026-10-14
Updated: 2026-10-14
Terminology
Sources
- Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents
- Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents
- Distilling the Knowledge in a Neural Network
- CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation
- Mistral 7B
- Compression-Aware Abstention: Teaching LLMs to Refuse When KV-Compression Masks Remove Answer Evidence
- Towards Reliable Context Compression for Long-Horizon Agents: An Empirical Study of Execution Instability
- Accuracy is Not Enough: A Divergence-Based Approach to Evaluate Fidelity Loss in Quantized LLMs
- Qwen2.5 Technical Report
- Compress the Context, Keep the Commitments: A Formal Framework for Verifiable LLM Context Compression
- Lost in Context Compaction: When LLM Systems Drop Side Constraints
- Less is More: DocString Compression in Code Generation
- LUMINA: Detecting Hallucinations in RAG System with Context-Knowledge Signals
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data