AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

RL-Guided Temporal Localization for Dual-Channel Retrieval in Long-Horizon Agent Memory

Dabin Sheng, Zhe Wu, Jinming Zhao, Junliang Xing, Yuanchun Shi

NeurIPS 2026 · Multi-agent systems and games › LLM agents · approach: Self-evolving language agent learning · 0 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 RL-guided temporal localization method with dual-channel retrieval for time-aware memory retrieval in long-horizon LLM agents.

Problem

Time-aware retrieval for long-horizon LLM agents over growing long-term memory; existing methods rely on lexical/semantic similarity and inject temporal cues only at organization or answer generation, so temporally mismatched but semantically similar memories crowd out targets.

Approach

Applies temporal constraints at the retrieval stage itself, using reinforcement learning to guide temporal localization within a dual-channel retrieval scheme (details beyond the abstract are not stated).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add the distance-decay reranker (ablation)TIDER · LifeBenchOverall accuracy (%): 57.60 -> 61.52Improved
Add the temporal channel (ablation)TIDER · LifeBenchOverall accuracy (%): 52.21 -> 57.60Improved
Add the timeline summary S_q (ablation)TIDER · LifeBenchOverall accuracy (%): 51.45 -> 52.21Improved
Add the timeline summary without a temporal channel (ablation)semantic-only retrieval · LifeBenchF1: 63.05 -> 47.62Hurt
Add GRPO to the SFT-initialized Qwen3-1.7B locator (ablation)TIDER locator · LifeBenchOverall accuracy (%): 55.64 -> 61.52Improved
Use a much larger zero-shot GPT-5.2 locator instead of the trained 1.7B locator (ablation)TIDER locator · LifeBenchOverall accuracy (%): 61.52 -> 57.84Hurt
Supply gold oracle windows instead of predicted windows (ablation)TIDER · LifeBenchOverall accuracy (%): +1.70%No clear effect
Supply oracle windows on LoCoMo (ablation)TIDER · LoCoMo TemporalTemporal accuracy (%): 71.65 -> 79.13Improved
Add detection and window-localization rewards in GRPO (ablation)TIDER · LifeBenchOverall accuracy (%): 55.88 -> 61.52Improved
Add outcome reward in GRPO (ablation)TIDER · LifeBenchOverall accuracy (%): 59.07 -> 61.52Improved
Add evidence reward in GRPO (ablation)TIDER · LifeBenchOverall accuracy (%): 59.56 -> 61.52Improved
Increase retrieval budget top-k from 20 to 100 (ablation)TIDER · LifeBenchOverall accuracy (%): +0.49%No clear effect

Limitations

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