AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

AssoMem: Scalable Memory QA with Multi-Signal Associative Retrieval

Kai Zhang, Xinyuan Zhang, Ejaz Ahmed, Hongda Jiang, Caleb Kumar, Kai Sun, Zhaojiang Lin, Sanat Sharma, Shereen Oraby, AARON COLAK, Ahmed A Aly, Anuj Kumar and 2 more

ICLR 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add temporal signal to the ranker (ablation)AssoMem · LongMemEval_mRecall@6: 73.39 -> 80.87Improved
add importance (PPR) signal to the ranker (ablation)AssoMem · LongMemEval_mRecall@6: 75.81 -> 80.87Improved
add clue nodes to the memory graph (ablation)AssoMem · LongMemEval_mRecall@6: 79.75 -> 80.87Improved
add clue nodes to the memory graph (ablation)AssoMem · LongMemEval_mAcc@10: 72.51 -> 69.17Hurt
use MI-driven weight assignment instead of fixed weights (ablation)AssoMem · LongMemEval_mRecall@6: 76.79 -> 80.87Improved
use mutual information fusion instead of learnable fusion (logistic regression) (ablation)AssoMem · LongMemEval_mRecall@6: 72.92 -> 82.64Improved
use mutual information fusion instead of information gain (ablation)AssoMem · LongMemEval_mRecall@6: 80.94 -> 82.64Improved
fine-tune generator with denoising multi-task data (ablation)Qwen2.5-32B-Instruct · LongMemEval_mAccuracy: 64.72 -> 73.88Improved
fine-tune generator with denoising multi-task data (ablation)LlaMA3.2-3B-Instruct · LongMemEval_mAccuracy: 26.91 -> 33.43Improved
use AssoMem retrieval instead of topic groupingbest baseline · LongMemEval_mRecall@10: 79.14 -> 84.96Improved
use AssoMem retrieval instead of topic grouping at larger memorybest baseline · LongMemEval_lRecall@10: 62.29 -> 69.33Improved
use hybrid multi-granularity retrieval instead of utterance-flatretrieval · LongMemEval_mRecall@10: 70.18 -> 78.97Improved

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