This paper shows that in the large depth and width asymptotics, Dropout and Random Gradient Masking (RaM) converge to the same limiting dynamics for ResNets.

The paper introduces 5WBENCH, a new benchmark for causal unlearning, and proposes MAAT, a novel three-phase framework that achieves high forgetting and high retention specifically on complex 'Why'-typ…

A simple online algorithm is presented for the stacking problem to avoid shifts with a sufficient condition involving stacking area dimension, load/unload points, and maximum items.

The paper introduces MemTrace, a framework that treats LLM memory pipelines as traceable graphs to systematically diagnose and automatically correct memory-related errors, boosting performance by up t…

This paper presents a memory-efficient training stack for Mixture-of-Experts (MoE) models, combining and specializing parallelism techniques for maximal efficiency.

This paper proposes a method to recover recoverability structure from failed traces of post-trained language models, enabling test-time routing and post-training analysis.

This paper systematically evaluates simple baseline approaches for using external memory (like SSDs) to solve difficult search problems, specifically focusing on the performance of immediate duplicate…

PURGE is a novel machine unlearning algorithm that leverages the duality between continual learning and unlearning to achieve high data retention while making the unlearned model indistinguishable fro…

The paper introduces two library combinators, prune and defrel/bank, to bring observational deduplication and bottom-up enumeration into the relational setting using plain miniKanren.

The paper proposes a novel, theoretically-grounded algorithm (HAMU) that addresses the challenge of machine unlearning by guaranteeing specified improvements in forget quality while minimizing retain…

The paper proposes methods for detecting training instability in large language models using internal monitors based on the functional role of critical modules and earliest computational sites.

The paper introduces AMNESIA, the first large-scale, open-source benchmark for medical unlearning, demonstrating that current unlearning methods struggle to separate individual patient data from share…

The paper presents a highly optimized, low-stack implementation of the HAETAE signature scheme, reducing peak stack usage significantly to enable its use on severely memory-constrained microcontroller…

This paper introduces a new protection code design for Content Addressable Memory (CAM) to ensure reliability.

The paper shows that deterministic cache eviction cannot ensure consistent serving-time error estimation and proposes a randomized approach to restore identifiability and provide error certificates.

The paper reframes context distillation as a latent memory management problem, proposing a modular framework using LoRA adapters and a Self-Gating mechanism for efficient, selective memory retrieval a…

This paper systematically studies memory poisoning attacks in LLM agents, identifying multiple vulnerabilities and proposing a new benchmark to assess the risk.

The paper introduces Sparse Memory-Efficient Training (SMET), a method that stabilizes and optimizes Dynamic Sparse Training (DST) for large language models, enabling stable and memory-efficient spars…

PoisonCap introduces a new 'poison' capability format for CHERI systems to provide efficient, strict use-after-free and initialization safety, surpassing existing temporal safety solutions.