Memory Is Not Always Needed: Characterizing Conditional Memory in Scientific Reasoning

• URL: http://arxiv.org/abs/2608.23982v1
• Date: Tue, 25 Aug 2026 02:19:20 GMT
• Title: Memory Is Not Always Needed: Characterizing Conditional Memory in Scientific Reasoning
• Abstract summary: We investigate when, where, and to what extent conditional memory should participate in scientific reasoning.<n>We propose a Knowledge Boundary-Aware Router that uses task-specific input proxies available before generation to determine whether memory is activated.<n>Our approach more consistently preserves beneficial memory contributions while suppressing memory-induced regressions.
• Score: 22.364424301865856
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Scientific reasoning requires language models to retrieve specialized knowledge and incorporate it reliably into multi-step computation. Conditional memory provides an explicit lookup pathway that complements dense neural representations, but its usefulness is inherently input- and computation-dependent: retrieved information may repair missing scientific associations, yet it may also introduce distracting shortcuts or interfere with reasoning that the base model can already perform correctly. In this work, we systematically investigate when, where, and to what extent conditional memory should participate in scientific reasoning. We characterize the scientific knowledge boundary and controlled interventions on memory-enabled knowledge-circuit nodes. Based on these analyses, we propose a Knowledge Boundary-Aware Router that uses task-specific input proxies available before generation to determine whether memory is activated, which layer-stage nodes receive memory signals, and how strongly these signals contribute. Experiments on biological and chemical reasoning benchmarks, covering two backbone families and six task types, show that memory effects vary substantially across inputs, tasks, and injection locations. Compared with static and activation-rate-matched random routing, our approach more consistently preserves beneficial memory contributions while suppressing memory-induced regressions, establishing selective memory allocation as an important principle for reliable scientific reasoning.

Related papers

• AI Engram: In Search of Memory Traces in Artificial Intelligence [19.383960470953145]
  This work introduces a geometric framework to identify such "AI engrams"<n>We derive a closed-form estimator that isolates individual memory traces from globally entangled parameters.<n>Results bridge theories of biological memory and artificial representation learning.
  arXiv  Detail & Related papers  (2026-06-12T22:36:52Z)
• Benchmarking local Hebbian learning rules for memory storage and prototype extraction [0.0]
  Associative memory or content-addressable memory is an important component function in computer science and information processing.<n>We benchmark associative memory function of seven different Hebbian learning rules employed in non-modular and modular recurrent networks.
  arXiv  Detail & Related papers  (2026-05-01T20:15:32Z)
• Evoking User Memory: Personalizing LLM via Recollection-Familiarity Adaptive Retrieval [59.295767860331004]
  RF-Mem is a familiarity uncertainty-guided dual-path memory retriever.<n>It embeds human-like dual-process recognition into the retriever.<n>It consistently outperforms one-shot retrieval and full-context reasoning under fixed budget and latency constraints.
  arXiv  Detail & Related papers  (2026-03-10T06:31:44Z)
• PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents [74.4491017652226]
  We propose PlugMem, a task-agnostic memory module that can be attached to arbitrary LLM agents without task-specific redesign.<n>Motivated by the fact that decision-relevant information is concentrated as abstract knowledge rather than raw experience, we draw on cognitive science to structure episodic memories into a compact, knowledge-centric memory graph.<n>We evaluate PlugMem unchanged across three benchmarks (long-horizon conversational question answering, multi-hop knowledge retrieval, and web agent tasks)
  arXiv  Detail & Related papers  (2026-02-06T18:55:38Z)
• AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agents [69.39123054975218]
  Memory serves as the pivotal nexus bridging past and future.<n>Recent research on autonomous agents has increasingly focused on designing efficient memory by drawing on cognitive neuroscience.
  arXiv  Detail & Related papers  (2025-12-29T10:01:32Z)
• Memory in the Age of AI Agents [217.9368190980982]
  This work aims to provide an up-to-date landscape of current agent memory research.<n>We identify three dominant realizations of agent memory, namely token-level, parametric, and latent memory.<n>To support practical development, we compile a comprehensive summary of memory benchmarks and open-source frameworks.
  arXiv  Detail & Related papers  (2025-12-15T17:22:34Z)
• MemGen: Weaving Generative Latent Memory for Self-Evolving Agents [57.1835920227202]
  We propose MemGen, a dynamic generative memory framework that equips agents with a human-esque cognitive faculty.<n>MemGen enables agents to recall and augment latent memory throughout reasoning, producing a tightly interwoven cycle of memory and cognition.
  arXiv  Detail & Related papers  (2025-09-29T12:33:13Z)
• Synaptic Theory of Chunking in Working Memory [0.5735035463793009]
  We introduce a synaptic theory of chunking, in which short-term synaptic plasticity enables the formation of chunk representations in working memory.<n>We show that a specialized population of chunking neurons'' selectively controls groups of stimulus-responsive neurons, akin to gating.<n>Our work provides a novel conceptual and analytical framework for understanding how the brain organizes information in real time.
  arXiv  Detail & Related papers  (2024-08-14T16:03:47Z)
• Sequential Memory with Temporal Predictive Coding [6.228559238589584]
  We propose a PC-based model for emphsequential memory, called emphtemporal predictive coding (tPC) We show that our tPC models can memorize and retrieve sequential inputs accurately with a biologically plausible neural implementation.
  arXiv  Detail & Related papers  (2023-05-19T20:03:31Z)
• Measures of Information Reflect Memorization Patterns [53.71420125627608]
  We show that the diversity in the activation patterns of different neurons is reflective of model generalization and memorization. Importantly, we discover that information organization points to the two forms of memorization, even for neural activations computed on unlabelled in-distribution examples.
  arXiv  Detail & Related papers  (2022-10-17T20:15:24Z)
• Entropic Associative Memory for Manuscript Symbols [0.0]
  Manuscript symbols can be stored, recognized and retrieved from an entropic digital memory that is associative and distributed but yet declarative. We discuss the operational characteristics of the entropic associative memory for retrieving objects with both complete and incomplete information.
  arXiv  Detail & Related papers  (2022-02-17T02:29:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.