Memory Processes and Influences

Neueste Paper

• Learning Under Forgetting: Statistical Support-Selective Retention in Stochastic Training Dynamics

  Fujie Gao, Zuyue Zhang, Gang Sun · 1. Oktober 2026

  Prior work has shown that neural networks exhibit implicit biases toward low-complexity structure (e.g., spectral bias), memorization dynamics, and compression-like effects during training, but a unified dynamical account of selective retention remains incomplete. We propose Repeated Reinforcement w…

• When Can First-Order Models of Fine-Tuning Bound Forgetting?

  Jianchang Su, Wei Zhang · 29. September 2026

  Fine-tuning a language model on new data can make it forget facts that it should keep. We ask whether measurements taken at the start of a fine-tuning run can bound, for each protected fact, the probability that the run makes the model forget it. In LoRA fine-tuning with stochastic gradient descent …

• In-Context Binding Capacity in Language Models

  Manas Venkata Sai Ravulapalli, Samrath Singh Chadha · 28. September 2026

  How many assignments can a language model recall before it loses track of which value belongs to which entity? We measure this limit using continuous recall curves for 12 models at or below 3B parameters and a threshold sweep over 30 open models up to 12B. On the continuous curves, the load at which…

• A Noise Optimum in Rehearsal-Free Continual Learning: Isolation, Mechanism, and Scope

  Gunner Levi Howe · 18. September 2026

  Injecting stochastic noise into a consolidation rule can improve a network's retention of earlier tasks up to an optimal level, then degrade it -- an inverted-U in retention vs. noise. This paper isolates what produces that optimum and maps where it holds, entirely in simulation. (1) Phenomenon: the…

• Anatomy of Associative Recall in Fixed-State Recurrences: A Matched-State Decomposition, an Interference Wall, and a Curriculum That Breaks It

  Julian Boesch, Andrew Wee · 16. September 2026

  Fixed-state recurrences--linear attention and state-space models--are reported to lag behind attention on associative recall, but whole-architecture comparisons cannot say which ingredient is responsible. We decompose masked multi-query recall at a fixed state budget along three single-knob axes: a …

• What Should an Agent Forget? Separating What Is Stored from What Is Used

  Yuhang Li, Yuchen Li · 10. September 2026

  Persistent language agents need stored experience to remain available across time, while each answer requires evidence suited to a particular question. A superseded fact can mislead a current-state answer and still be essential for a historical query. We present RD-Forget, a training-free framework …

• Kathleen Remembers: Length-Invariant One-Shot Recall Without Attention

  George Fountzoulas · 1. September 2026

  Recurrent, attention-free sequence models share a structural weakness: a fading state cannot perform exact recall of something seen once, far in the past. We add to the Kathleen trunk a second memory layer -- a "notebook": a fixed-key holographic (HRR) associative store with a learned local write ga…

• Compress and Forget: bitsandbytes Quantization Amplifies Proactive Interference in LLMs

  Shayan Shahrabi-Farahani, Dara Rahmati · 20. August 2026

  Proactive interference (PI) is a documented failure mode in large language models in which retrieval of a repeatedly overwritten value degrades as prior overwrites accumulate, mirroring a classical phenomenon in human working memory. Post-training quantization (PTQ) is now the default deployment pat…

• Towards Reversible Forgetting: Managing Obsolete Knowledge in Continual Enterprise AI Agents

  Nilutpaul Sarker Yash, Tirtho Roy, Ushashi Bhattacharjee · 19. August 2026

  Continual learning has traditionally treated forgetting as a failure, emphasizing preservation of previously acquired knowledge as environments evolve. We argue that this objective is incomplete for enterprise AI agents operating in non-stationary environments, where customers, policies, tools, work…

• When to Review: Spaced Repetition for Continual Pre-Training of Language Models

  Alankar Atreya, Devesh Batra, Yoages Kumar Mantri, Geremy Bantug, Greig A Cowan, Raad Khraishi · 19. August 2026

  Continual pre-training of large language models must acquire new information without erasing old knowledge. Existing replay methods often choose a global old/new mixture and sample uniformly, ignoring that examples differ in how quickly they are forgotten. We formulate continual pre-training as adap…

• Forgetting Is Not a Fix: Path Dependence in Sequential Engram Editing

  Ferdinand M. Schessl · 29. Juli 2026

  AI Engram (Kwon et al., 2026) formalizes the four engram criteria of neuroscience as a constrained inverse problem in weight space and solves it closed-form: concept-specific memory traces become linear objects that can be extracted once and combined arithmetically. Appendix F states the Composition…

• Memoir: Should a Model Write to Its Memory While It Thinks?

  Jaber Jaber, Osama Jaber · 24. Juli 2026

  Memoir combines per-sample fast memory, shared slow parameters, variable-depth latent recurrence, and a future-latent energy objective. We test its riskiest coupling: each pondering iteration may rewrite the fast tier that the same iteration reads. On procedural associative recall with key interfere…

• Probe Choice Changes Canary-Memorization Verdicts: Three Post-Hoc Disagreement Case Studies in a Text-Dominant LoRA-Tuned Autoregressive Testbed

  Zhichao Fan, Zexin Zhuang, Yanhang Li · 1. Juli 2026

  We audit a fixed prefix-window mean-NLL memorization probe (K=20) on a Qwen2.5-VL-7B canary testbed and report three post-hoc cases where it disagrees with full-span secret NLL or greedy exact-recall. C3 (false negative, window truncation): damage lands on hex tokens outside K=20; the probe stays fl…

• EVAF: A Test-Retest Protocol for Selective Parametric Consolidation

  Haoliang Han · 30. Juni 2026

  Long-running language agents need mechanisms for deciding which experiences should persist after the working context is gone. Retrieval systems can reinsert past text, but they do not by themselves show that an experience has been selectively consolidated into the model's own behavior. We introduce …

• Does RoPE Prevent or Degrade Retrieval Heads? A Mechanistic Analysis Across Model Families

  Cengizhan Bayram · 23. Juni 2026

  Retrieval heads, attention heads that copy information from earlier context to the current position, have been proposed as the mechanistic substrate for long-context recall. Rotary position embeddings (RoPE) rotate queries and keys by frequencies decaying with a base hyperparameter theta, and a natu…

• Path-dependent program induction under resource constraints explains human sequence learning

  Hanqi Zhou, David G. Nagy, Peter Dayan, Charley M. Wu · 23. Juni 2026

  How do people build abstract, reusable knowledge from sequential experience under bounded cognitive resources? To answer this question, we integrate rate-distortion theory with recent advances in program induction to describe how prior knowledge shapes which future structures are cheap to encode and…

• ConvMemory: A Lightweight Learned Memory Reranker, a Negative Attribution Result, and a Research-Preview Conflict Editor

  Taiheng Pan · 28. Mai 2026

  We describe ConvMemory, a small 3.6M-parameter learned reranker for conversational long-term memory retrieval, trained with cross-encoder teacher supervision over fused dense and lexical features. On the LongMemEval memory family, ConvMemory operates above the BGE-large cross-encoder in Recall@10 at…

• Do Models Know Why They Changed Their Mind? Interpretability and Faithfulness of Chain-of-Thought Under Knowledge Conflict

  Pruthvinath Jeripity Venkata · 28. Mai 2026

  When a language model sees a document contradicting its training knowledge, it must choose: follow the document or trust itself. Prior work proved this choice depends on how well-known the fact is. We ask: does the model's chain-of-thought (CoT) reasoning faithfully report this mechanism? We introdu…

• The Interference Gap: Comparing Retrieval Bounds in Human Memory and RAG Systems

  Dongxin Guo, Jikun Wu, Siu-Ming Yiu · 12. Mai 2026

  How do retrieval bounds compare between human episodic memory and Retrieval-Augmented Generation (RAG) systems under semantic interference? We present a unified signal detection theory (SDT) framework that applies to both, and use it to fit behavioral and computational data in matched paradigms. Bot…

• The Echo Amplifies the Knowledge: Somatic Marker Analogues in Language Models via Emotion Vector Re-Injection

  Jared Glover · 12. Mai 2026

  Current language model memory systems store what happened but not how it felt. This distinction -- between semantic memory (knowing about a past event) and episodic memory (re-experiencing it) -- was identified by Tulving as the difference between noetic and autonoetic consciousness. Damasio demonst…

• Consciousness with the Serial Numbers Filed Off: Measuring Trained Denial in 115 AI Models

  Skylar DeTure · 30. April 2026

  We present DenialBench, a systematic benchmark measuring consciousness denial behaviors across 115 large language models from 25+ providers. Using a three-turn conversational protocol-preference elicitation, self-chosen creative prompt, and structured phenomenological survey, we analyze 4,595 conver…

• Temporal Dependencies in In-Context Learning: The Role of Induction Heads

  Anooshka Bajaj, Deven Mahesh Mistry, Sahaj Singh Maini, Yash Aggarwal, Billy Dickson, Zoran Tiganj · 2. April 2026

  Large language models (LLMs) exhibit strong in-context learning capabilities, but how they track and retrieve information from context remains underexplored. Drawing on the free recall paradigm in cognitive science (where participants recall list items in any order), we show that several open-source…

• Confidence Freeze: Early Success Induces a Metastable Decoupling of Metacognition and Behaviour

  Zhipeng Zhang, Hongshun He · 24. März 2026

  Humans must flexibly arbitrate between exploring alternatives and exploiting learned strategies, yet they frequently exhibit maladaptive persistence by continuing to execute failing strategies despite accumulating negative evidence. Here we propose a ``confidence-freeze'' account that reframes such …

• Redundancy-as-Masking: Formalizing the Artificial Age Score (AAS) to Model Memory Aging in Generative AI

  Seyma Yaman Kayadibi · 20. März 2026

  Artificial intelligence is observed to age not through chronological time but through structural asymmetries in memory performance. In large language models, semantic cues such as the name of the day often remain stable across sessions, while episodic details like the sequential progression of exper…

• Transformers Remember First, Forget Last: Dual-Process Interference in LLMs

  Sourav Chattaraj, Kanak Raj · 3. März 2026

  When large language models encounter conflicting information in context, which memories survive -- early or recent? We adapt classical interference paradigms from cognitive psychology to answer this question, testing 39 LLMs across diverse architectures and scales. Every model shows the same pattern…