cs.CLAug 18, 2026

CABLE: Extending the Reach of Memory Retrieval via Complementary Antecedent-Based Linking and Expansion

Authors: Zheling Tan, Jin Gao, Dequan Wang

Organizations: Shanghai Jiao Tong University · Shanghai Innovation Institute

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

As LLM agents operate across structured workflows and sessions, preserving long-term history does not ensure that later contexts can recover relevant evidence through a bounded memory interface. We study this evidence-reachability problem in long-term conversational memory, where retrieval still relies heavily on semantic similarity. This works well for topical recall, but it often misses earlier experiences, plans, or motivations that are semantically distant from the later events they help explain. Existing memory graphs provide cross-memory structure, yet links driven mainly by semantic overlap can duplicate what the host retriever already recovers. We argue that link construction should instead prioritize a sparse set of retriever-complementary associations. We present CABLE (Complementary Antecedent-Based Linking and Expansion), a plug-in augmentation that constructs links designed to extend the host retriever's direct semantic reach. For each new memory, CABLE generates antecedent-oriented queries, retrieves prior memories, subtracts candidates in the direct semantic neighborhood, and verifies the remainder before adding the accepted complementary associations into a sparse directed graph. At retrieval time, CABLE expands the host system's retrieved seeds along these links to surface implicit supporting evidence. We evaluate CABLE with A-MEM on LoCoMo and MA-LongMemEval, and further integrate it into SimpleMem and Mem0g on LoCoMo, using Qwen3.5-27B, DeepSeek-chat, and GPT-4o-mini. CABLE yields higher mean LLM-judge scores in every evaluated system-level setting, with the largest gains in categories where useful evidence is distributed across memories or sessions, including open-domain, multi-session, and preference-oriented questions. These results support prioritizing sparse, reasoning-relevant associations that complement rather than duplicate the host retriever.

Figures & tables

Figure 1: Stored evidence can remain outside direct retrieval. Direct semantic retrieval returns Memory B but not Memory A, which contains the required answer. During construction, CABLE stores the antecedent link ( A→BA\rightarrow B ); at query time, expansion starts from the retrieved Memory B and follows its incoming link to recover Memory A.
Figure 2: CABLE constructs sparse antecedent links and reuses them for bounded retrieval expansion. (A) Direct semantic retrieval reaches Memory B but misses the semantically distant yet relevant Memory A. (B) During construction, CABLE retrieves direct matches ( BiB_{i} ) and antecedent candidates ( HiH_{i} ), subtracts their overlap, verifies the remaining candidates, and stores accepted links ( mj→mim_{j}\rightarrow m_{i} ) in graph G. (C) At inference, the host system returns ( R0R_{0} ), CABLE selects reliable seeds ( SS ), expands their stored graph neighborhoods, and applies candidate scoring and novelty filtering. Similarity values are illustrative.
BenchmarkModelBaseline+CABLEΔ\Delta Score (pp)
LoCoMoQwen3.5-27B71.2374.81+3.58
 DeepSeek-chat68.1570.26+2.11
MA-LongMemEvalQwen3.5-27B59.3365.33+6.00
 GPT-4o-mini48.6749.67+1.00
Table 1: CABLE improves A-MEM in all four settings, with gains ranging from 1.00 to 6.00 percentage points. Scores are mean LLM-judge percentages; Δ\Delta Score denotes +CABLE minus the matched baseline.
BackboneCategoryBaseline+CABLEΔ\Delta Score (pp)
Qwen3.5-27BSingle-hop74.0876.81+2.73
 Multi-hop73.7676.95+3.19
 Temporal68.8574.14+5.29
 Open-domain46.8853.12+6.24
DeepSeek-chatSingle-hop73.9675.51+1.55
 Multi-hop63.6565.25+1.60
Table 2: CABLE improves every LoCoMo category under both backbones, with the largest gains on open-domain questions. Values are mean LLM-judge scores (%).
Figure 3: Category-wise A-MEM results on MA-LongMemEval. CABLE improves multi-session and single-session-preference under both backbones, but lowers temporal-reasoning. Labels report changes from baseline; higher is better.
SystemModelBaseline+CABLEΔ\Delta Score (pp)
SimpleMemQwen3.5-27B81.6982.27+0.58
 DeepSeek-chat79.4281.04+1.62
Mem0 gQwen3.5-27B52.6054.80+2.20
Table 3: CABLE yields higher mean scores for adaptive-retrieval SimpleMem and graph-based Mem0g on LoCoMo, indicating that retriever complementarity is not specific to A-MEM. Comparisons are made within rows; Δ\Delta Score is reported in percentage points.
Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

HostHost retrievalCABLE activationBudget control
A-MEMDirect host retrievalCABLE expands after host retrieval and replaces entries only when verified, novel candidates are available.Both systems pass at most 45 entries to the answer generator. Up to five CABLE candidates replace the five lowest-ranked baseline entries.
SimpleMemHost-controlled adaptive retrieval with an LLM reflection stepCABLE is activated only when the reflection step judges the baseline retrieval insufficient.SimpleMem retains its native adaptive protocol; no fixed cross-system entry budget is imposed.
Mem0 gGraph-enabled host retrieval over extracted entities and relationsCABLE expands after the host graph retrieval.Both systems use a fixed 20-entry budget. Up to five CABLE candidates replace baseline entries.
Appendix
Table 4: Integration protocols preserve each host system’s retrieval regime. A-MEM and Mem0 g use matched fixed retrieved-entry budgets, whereas SimpleMem retains adaptive retrieval and invokes CABLE only after insufficient baseline retrieval.
VariantLoCoMoΔ\DeltaMA-LongMemEvalΔ\Delta
Full CABLE74.81–65.33–
w/o overlap subtraction73.90−0.91-0.9164.66−0.67-0.67
w/o verification74.61−0.20-0.2062.67−2.66-2.66
Generic query decomposition74.35−0.46-0.46––
Appendix
Table 5: Removing overlap subtraction lowers performance on both benchmarks, while removing verification causes the largest degradation on MA-LongMemEval. Generic query decomposition also underperforms the type-conditioned antecedent queries on LoCoMo. Results use A-MEM with Qwen3.5-27B; scores are mean LLM-judge percentages, and Δ\Delta is relative to full CABLE in percentage points. A dash denotes an unevaluated setting.
ItemEvidence
QuestionWhy did Melanie choose to use colors and patterns in her pottery project?
Ground truthShe wanted to catch the eye and make people smile.
Direct-retrieval seedD5:6 : “I’m a big fan of pottery – the creativity and skill is awesome. Plus, making it is so calming. Look at this!”
Stored CABLE linkD5:6 →\rightarrow D12:6
CABLE-expanded evidenceD12:6 : “Thanks, Caroline! I’m obsessed with those, so I made something to catch the eye and make people smile. Plus, painting helps me express my feelings and be creative. Each stroke carries a part of me.”
AnswersBaseline: “Not mentioned in context.” +CABLE: “To catch the eye, make people smile, and express her feelings creatively.”
Appendix
Table 6: A concrete evidence-reachability failure and its recovery. Direct retrieval returns a topically similar pottery memory that lacks the required motivation and yields an abstaining answer. CABLE expands through a stored antecedent link, recovers the otherwise unretrieved motivation in D12:6, and produces a reference-supported answer.
More in this topic

Conversational Memory

Conversational memory equips AI assistants to retain, update, and retrieve useful information from past interactions for coherent, personalized dialogue.

See all 71 papers

Explore similar work

CardsList
Aug 13, 2026

RippleMem: From Isolated Retrieval to Associative Recollection for Long-Term Agent Memory

LLM-based agents increasingly rely on external memory to support long-horizon reasoning and interaction. However, the main bottleneck is not simply storing past experience, but recovering the right set of evidence when relevant information is distributed across many interactions. Existing approaches struggle with this access problem. Full-context methods require noisy long-context search, flat retrieval often returns isolated and incomplete records, and graph-based memory systems can be expensive to construct while compressing rich event context. We introduce RippleMem, a long-term memory system that replaces one-shot retrieval with adaptive associative recollection. Inspired by cue-dependent episodic retrieval and associative completion, RippleMem stores interaction history as cue-rich episodic memory units and organizes them in an event-centric memory graph. Given a query, it first recalls relevant memory anchors through hybrid cues, then expands from these anchors along semantic and structural associations to recover missing supporting evidence. In this way, initially recalled memories serve not only as answer context, but also as cues for completing the evidence needed to answer. Experiments on LoCoMo and LongMemEval-S show that RippleMem achieves the best overall performance across evaluated settings, improving LLM-as-a-Judge accuracy by 3.95% on LoCoMo and up to 11.87% on LongMemEval-S, while reducing graph construction cost by about 30x.

Jingbo Ji, Lingyi Li, Xilong Cheng +4

Communication University of China, Beijing, China. · Zhilian Yinghe Technology Co., Ltd., Beijing, China · State Key Laboratory of Media Convergence and Communication, Beijing, China

Agent MemoryPersistent Memory for Language ModelsGraph-Based Agent MemoryAssociative Memory
AskPDF
Oct 15, 2026

From Relevant Memories to Sufficient Evidence: Rethinking Memory Utilization in LLM Agents

Long-term memory enables LLM agents to reuse past experiences, yet retrieving relevant memories does not necessarily provide sufficient evidence for answering a query. Essential information may remain fragmented across memory records, obscured by abstraction, or disconnected from the relations required for reasoning. We identify this relevance-sufficiency gap and propose QEMem, a framework that reformulates memory utilization as the construction of query-specific evidence sets, shifting the focus from individual retrieval relevance to collective evidential support. QEMem integrates three complementary mechanisms: query-guided discovery retrieves evidence across multiple memory granularities; source-grounded reification organizes it into structured ViewCards with explicit evidential roles and source references; and sufficiency-guided refinement selectively recovers missing information. Operating on existing memory stores without modifying memory formation, QEMem achieves 91.43% LLM-judge accuracy on LoCoMo with Claude Opus 4.5, outperforming the strongest evaluated baseline, MemR3^3, by 7.6 percentage points and NEMORI by 10.7 points under a shared memory store. Further analyses demonstrate improved evidence coverage and strong performance on questions requiring cross-session evidence integration.

Shixuan Li, Wei Yang, Peiyu Zhang +7

University of Southern California · Georgia Institute of Technology

LLM Agent MemoryLLM AgentsRetrieval-Augmented Reasoning
AskPDF
Jun 22, 2026

RaMem: Contextual Reinstatement for Long-term Agentic Memory

Long-term memory has become increasingly important for LLM agents that operate across extended interactions and evolving task contexts. Recent memory systems have made past experiences more persistent, compact, and retrievable, but retrieval alone does not ensure that a memory provides valid evidence for the current query. When experiences are compressed into reusable fragments, memories from different situations may appear equally relevant if they involve recurring entities or user states. We refer to this failure as context collapse: memories lose the surrounding context needed to judge whether they provide valid evidence for the current query. To address this problem, we propose Contextual Reinstatement for Agentic Memory (RaMem), a framework that turns retrieved memory fragments into contextually verifiable evidence. RaMem operates through four coordinated stages: (i) evidence anchoring grounds each memory in its original episodic conditions, especially event time, mention time, session span, and participants; (ii) recall condition induction derives the evidence conditions implied by the query; (iii) validity-aware retrieval uses these conditions to prioritize context-compatible memories while retaining content-relevant candidates as fallback evidence; and (iv) context-preserved synthesis keeps the selected memories' structured context available to the generator. Experiments on long-term memory benchmarks show that RaMem consistently improves performance over strong memory baselines, with average F1 gains of more than 10% across several backbones.

Wei Yang, Bryce Kan, Shixuan Li +5

University of Southern California

Agent MemoryLLM Agent Memory
AskPDF