SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2607.13591 · 2026

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents

Kai-Wei Chang, Ying Wu, Dong Liu, Xiao Liang, Zhi Zhang, Haozheng Luo, Edward Sun, Los Angeles, Rui Sun, Zhaolu Kang, Aylin Caliskan, Yuchen Wu, and 3 more

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2607.13591.json · badge: /api/ran/2607.13591.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ericjiang18/MemCon canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
FunctionStatusWhere it lives
fmt_acc Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/agent_baseline/summarize_results.py
pointer only (licence: NONE) · get_code("fe7396233e692e2a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

fmt_tok Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/agent_baseline/summarize_results.py
pointer only (licence: NONE) · get_code("314b64d7931e49bf")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_model_type Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/tasks/utils.py
pointer only (licence: NONE) · get_code("22fbf3c1d8c1b401")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_config Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/mas/llm.py
pointer only (licence: NONE) · get_code("e266c82417005686")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_json Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/mas/utils.py
pointer only (licence: NONE) · get_code("fb7d225e5f4b88c5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_results Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/agent_baseline/summarize_results.py
pointer only (licence: NONE) · get_code("53e1ee4507b2f389")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

random_divide_list Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ericjiang18/MemCon/mas/utils.py
pointer only (licence: NONE) · get_code("b31f5961a38c2ccb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

resolve_benchmark Ran
ran; our contract draft was wrong, not the code
ericjiang18/MemCon/agent_baseline/run_benchmark.py
pointer only (licence: NONE) · get_code("e11b2ba100fa22aa")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ericjiang18/MemCon when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 0 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
fmt_acc 1 other paper
1 dated earlier
2604.00344 (2026-04)
Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
fmt_tok 1 other paper
1 dated earlier
2604.00344 (2026-04)
Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
load_results 1 other paper
1 dated earlier
2604.00344 (2026-04)
Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
resolve_benchmark 1 other paper
1 dated earlier
2604.00344 (2026-04)
Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present MEMCON (Memory as a Controlled Process), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MEMCON is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MEMCON consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5-20%. Our code is available at https://github.com/ericjiang18/MemCon/

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2607.13591")
get_code_for_paper("2607.13591")
have("2607.13591")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2607.13591.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2607.13591.svg)](https://syntology.ai/paper/2607.13591)

Connect an agent — have() is free.