AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Knowing What is Missing: Efficient Conversational Memory via Explicit Evidence-Gap Tracking

Xingbo Du, Loka Li, Duzhen Zhang, Leonard Song, Le Song

NeurIPS 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning · 36 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes an efficient conversational memory system for LLM agents that explicitly tracks evidence gaps to guide multi-step retrieval over long dialogue histories, aiming to avoid context expansion and token cost.

Problem

Long-term conversational memory for LLM agents: full-history reasoning is token-expensive and noisy, while retrieval-only methods fail on multi-hop and temporal questions; existing multi-step retrievers grow the textual context over time.

Approach

Explicitly tracks which evidence is still missing for a question and uses those gaps to drive retrieval steps, instead of appending ever-growing raw textual history to the context. Details beyond the abstract are not available.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
wrap RAG backend with MemR3 evidence-gap controllerRAG · LoCoMo overall (GPT-4o-mini)LLM-judge (%): 75.54 -> 81.55Improved
wrap RAG backend with MemR3RAG · LoCoMo overall (GPT-4.1-mini)LLM-judge (%): 79.46 -> 86.75Improved
wrap Zep backend with MemR3Zep · LoCoMo overallLLM-judge (%): 74.62 -> 76.26 (4o); 78.94 -> 80.88 (4.1)Improved
add MemR3 to RAGRAG · LoCoMo temporal (GPT-4o-mini)LLM-judge (%): 65.11 -> 76.22Improved
add absorb-and-discard masking of raw snippets (ablation)MemR3 (RAG) · LoCoMo overallLLM-judge (%): 68.54 -> 81.55Improved
add state-driven refinement query delta q_k (ablation)MemR3 (RAG) · LoCoMo overallLLM-judge (%): 77.11 -> 81.55Improved
add reflect action (ablation)MemR3 (RAG) · LoCoMo overallLLM-judge (%): 76.65 -> 81.55Improved
use implicit multi-step retrievers (Self-RAG, RCR) instead of vanilla RAGRAG · LongMemEval_saccuracyHurt
use MemR3 instead of Full-ContextLLM backends · LongMemEval_saccuracyMixed
use MemR3 instead of Full-ContextGPT-4o-mini · LongMemEval_s latencyavg latency (s): 11.04 -> 3.87Improved
replace model-estimated state with human-verified oracle state (ablation)MemR3 · LoCoMo 152-question audithuman accuracy (%): 63.2 -> 84.2Improved
MemR3 versus Full-Context on LoCoMo overallMemR3 (RAG) · LoCoMo overall (GPT-4.1-mini)LLM-judge (%): 86.75 vs 89.00Hurt

Experimental setup

Baselines: multi-step retrievers

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy