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

• URL: http://arxiv.org/abs/2607.24805v1
• Date: Mon, 06 Jul 2026 23:45:06 GMT
• Title: Forgetting Is Not a Fix: Path Dependence in Sequential Engram Editing
• Abstract summary: We test the hypothesis that concept-specific memory traces become linear objects that can be extracted once and combined arithmetically.<n>We run that test on the authors' own reference implementation, at their reported best edit strength.<n>For unlearning-as-compliance: erasure certified today does not certify the artifact after its next edit.
• Score: 35.76482964927589
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 Compositional Memory States Hypothesis: edited models live on "a commutative manifold where the integration of A and B reaches a consistent equilibrium regardless of the learning sequence." The evidence base is single and paired edits -- in materials terms, single-cycle tests, in which fatigue accumulation is structurally invisible. Whether the hypothesis holds under sequential load is exactly the "temporal dynamics" question the paper defers to future work. We run that test on the authors' own reference implementation, at their reported best edit strength (TOFU alpha=0.6, a choice favoring the linearity hypothesis), with pre-registered predictions, across three model charges (two vendors, two architecture families). Four findings replicate across all three: (1) zero-shot composition and sequential re-calibrated editing diverge by 61-71% of the edit magnitude; (2) cut order is not interchangeable, and the effect scales with concept overlap -- in one charge the order of cutting two Paris landmarks decides whether an uninvolved third concept survives; (3) the survivors' layer-input covariances -- the method's own sufficient statistics, read as strain gauges -- drift monotonically with every further cut, in every surviving concept, in every charge; (4) erased knowledge partially returns under subsequent unrelated cuts. Appendix F's commutative-manifold hypothesis is thereby falsified for sequential editing; the single-edit results of the original paper are untouched. For unlearning-as-compliance: erasure certified today does not certify the artifact after its next edit.

Related papers

• Calibrating Interpretability Instruments Before Trusting Their Verdicts [1.0829694003408499]
  Causal claims about large language model (LLM) internals rest on measurements.<n>These measurements fail in specific, diagnosable ways that return a plausible number instead of an error.<n>This note documents six such failures from a causal interpretability program on refusal and moral representation.
  arXiv  Detail & Related papers  (2026-09-13T19:37:02Z)
• Feature Priming in Online Linear Regression: Sparse-Regret Lower Bounds and a Tight Univariate Rate [47.36630149538994]
  In high-dimensional online prediction, the best predictor may depend on only a few features, so regret should scale with sparsity rather than the ambient dimension.<n>Warmuth and Amid asked at COLT 2023 whether any of three such rules admits a competitive online regret guarantee.<n>Our analysis identifies a common obstruction: cheap nuisance causes the refit to underweight the truly predictive coordinate.
  arXiv  Detail & Related papers  (2026-08-18T09:32:03Z)
• A Calibrated Test of Internal Action Maps: State Signals Without Global Affine Closure [0.16921396880325776]
  We test whether action maps fitted without a source reach its natural post-action activation and compose.<n>We organize the tests as an evidence lattice and validate the geometric branch on a known affine S_5 carrier.
  arXiv  Detail & Related papers  (2026-08-13T05:41:24Z)
• One Recipe, Many Harnesses: What Self-Evolution Encodes Across Languages and Models [46.83445434865936]
  Self-evolving harnesses are closed-loop systems in which an agent inspects its own rollouts and edits its prompts, tools, and memory.<n>It remains unclear whether they encode benchmark-specific adaptations, language-specific engineering knowledge, or compensation for limitations of the underlying model.
  arXiv  Detail & Related papers  (2026-08-10T19:45:45Z)
• Consistency Has a Computable Blind Spot: A Commutation Theory of Label-Free Reliability for Vision-Language Figure Reading [0.8594140167290097]
  We act on the complementary relation, equivariance: edit a figure's data and the correct answer must change by a computable amount.<n>Two matched edits are provably complete for affine reading errors; no suite of swap edits is complete for label permutations, and cyclic relabeling closes most of that gap.<n>The predicted ordering holds across three models and a hand-labeled population immune to the one circularity in how it is selected.
  arXiv  Detail & Related papers  (2026-08-06T07:14:58Z)
• DriftScope: Measuring The Hidden Effects of Diffusion Model Adaptation [50.584834483403675]
  Adapting pre-trained text-to-image diffusion models, whether to learn new visual concepts or erase unwanted ones, is routinely evaluated on its intended effects alone.<n>We demonstrate that adaptation systematically damages semantically unrelated concepts in ways that aggregate metrics structurally cannot surface.<n>We introduce DriftScope, a prompt-level diagnostic tool that takes any two model checkpoints and returns a ranked list of tokens whose visual concepts have shifted most between them.
  arXiv  Detail & Related papers  (2026-06-30T20:57:51Z)
• The Deterministic Horizon: Impossibility Results as Design Specifications for Trustworthy AI Systems [0.0]
  This thesis turns impossibility results from curiosities into design rules.<n>Its flagship result proves an accuracy ceiling set by architecture alone.<n>The same argument recasts across subfields.
  arXiv  Detail & Related papers  (2026-05-21T20:48:35Z)
• Three Regimes of Context-Parametric Conflict: A Predictive Framework and Empirical Validation [0.0]
  We study how large language models handle conflict between their training knowledge and a contradicting document.<n>We propose a three-regime framework: Regime 1 (single-source updating), Regime 2 (competitive integration), and Regime 3 (task-appropriate selection)<n>We validate the framework across Claude Sonnet 4.6, GPT-5.5, Gemini 2.5 Flash, Llama 4 Maverick, and DeepSeek V3 using 9,970 API calls in three experimental phases.
  arXiv  Detail & Related papers  (2026-05-12T06:00:48Z)
• The Digital Twin Counterfactual Framework: A Validation Architecture for Simulated Potential Outcomes [0.0]
  Digital Twin Counterfactual Framework (DTCF)<n>Five-level validation architecture converts unfalsifiable claim that simulator produces correct counterfactuals into falsifiable tests against observable data.<n> decomposition separates causal quantities into those that are marginally validated.<n>Joint causal claims become explicitly assumption-indexed.
  arXiv  Detail & Related papers  (2026-04-01T19:04:15Z)
• Causal Identification from Counterfactual Data: Completeness and Bounding Results [54.147490305295456]
  We develop an algorithm for identifying counterfactual queries from an arbitrary set of Layer 3 distributions.<n>We establish the theoretical limit of which counterfactuals can be identified from physically realizable distributions.<n>We derive novel analytic bounds for such quantities using realizable counterfactual data.
  arXiv  Detail & Related papers  (2026-02-26T22:48:12Z)
• The Mirror Loop: Recursive Non-Convergence in Generative Reasoning Systems [0.0]
  Recursive self-evaluation without external feedback often yields reformulation rather than progress.<n>We study 144 reasoning sequences across three models (OpenAI GPT-4o-mini, Anthropic Claude 3 Haiku, and Google Gemini 2.0 Flash) and four task families (arithmetic, code, explanation, reflection)<n>We interpret this as evidence for a structural limit on self-correction in generative reasoning.
  arXiv  Detail & Related papers  (2025-10-23T07:53:26Z)
• Neuro-Symbolic Entropy Regularization [78.16196949641079]
  In structured prediction, the goal is to jointly predict many output variables that together encode a structured object. One approach -- entropy regularization -- posits that decision boundaries should lie in low-probability regions. We propose a loss, neuro-symbolic entropy regularization, that encourages the model to confidently predict a valid object.
  arXiv  Detail & Related papers  (2022-01-25T06:23:10Z)
• Nested Counterfactual Identification from Arbitrary Surrogate Experiments [95.48089725859298]
  We study the identification of nested counterfactuals from an arbitrary combination of observations and experiments. Specifically, we prove the counterfactual unnesting theorem (CUT), which allows one to map arbitrary nested counterfactuals to unnested ones.
  arXiv  Detail & Related papers  (2021-07-07T12:51:04Z)
• L2R2: Leveraging Ranking for Abductive Reasoning [65.40375542988416]
  The abductive natural language inference task ($alpha$NLI) is proposed to evaluate the abductive reasoning ability of a learning system. A novel $L2R2$ approach is proposed under the learning-to-rank framework. Experiments on the ART dataset reach the state-of-the-art in the public leaderboard.
  arXiv  Detail & Related papers  (2020-05-22T15:01:23Z)

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.