Persistent Context Graphs for Efficient Memory Compaction in LLM Agents

A support or data assistant repeats work across related customer requests.

What you would change

Try the paper's memory change in a replay of past tasks, with separate memory per customer.

Try the paper's memory change in a replay of past tasks, with separate memory per customer.

Compare against

Your current memory pipeline plus a no-memory control using the same agent.

1. Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.
2. Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.
3. Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.
4. Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test.

What to measure

Task success, stale memories, retrieval misses, response latency, and all memory maintenance costs.

When to keep it

Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation.

recommendation

test

research area

memory and context

technical change

Attention-derived persistent context graph for memory compaction

published

2026-09-30

The method is implementable on a single GPU without training, but its dependence on attention access and its demonstrated gains being limited to specific agentic coding benchmarks require workload-specific validation before production adoption.

1. claim 01Paperraft inference
   Read the brief interpretation

   ReCAP replaces summarization-based and KV-cache-based history compaction with a persistent context graph of attention-derived importance scores and dependency links, so message selection for each new request uses stored signals plus lightweight relevance matching against the request rather than additional model calls. The cost is instrumentation to capture and store attention statistics and graph links per message, plus implementation complexity, and it presumes access to attention weights, which third-party APIs generally do not expose. It can fail when attention-based importance diverges from actual task relevance, when dependency links omit context a later request needs, or when the latency estimates do not transfer to the reader's real serving stack and workloads.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source
2. claim 02Paperraft inference

   Reduces estimated latency for compaction and cold restoration by approximately 95% versus Codex's summarization-based compaction on Qwen3-Coder and gpt-oss; roughly halves historical context per call on SWE-Together at comparable quality; improves accuracy on Lost-in-Conversation code tasks by 19.8 and 41.2 points over full history.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source