ROKA: Robust Knowledge Unlearning against Adversaries
arXiv:2603.00436v1 Announce Type: new Abstract: The need for machine unlearning is critical for data privacy, yet existing methods often cause Knowledge Contamination by unintentionally damaging related knowledge. Such a degraded model performance after unlearning has been recently leveraged for new...
TRIZ-RAGNER: A Retrieval-Augmented Large Language Model for TRIZ-Aware Named Entity Recognition in Patent-Based Contradiction Mining
arXiv:2602.23656v1 Announce Type: new Abstract: TRIZ-based contradiction mining is a fundamental task in patent analysis and systematic innovation, as it enables the identification of improving and worsening technical parameters that drive inventive problem solving. However, existing approaches largely rely on...
GLUScope: A Tool for Analyzing GLU Neurons in Transformer Language Models
arXiv:2602.23826v1 Announce Type: new Abstract: We present GLUScope, an open-source tool for analyzing neurons in Transformer-based language models, intended for interpretability researchers. We focus on more recent models than previous tools do; specifically we consider gated activation functions such as...
The Astonishing Ability of Large Language Models to Parse Jabberwockified Language
arXiv:2602.23928v1 Announce Type: new Abstract: We show that large language models (LLMs) have an astonishing ability to recover meaning from severely degraded English texts. Texts in which content words have been randomly substituted by nonsense strings, e.g., "At the ghybe...
CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning
arXiv:2602.24142v1 Announce Type: new Abstract: Mobile Agents can autonomously execute user instructions, which requires hybrid-capabilities reasoning, including screen summary, subtask planning, action decision and action function. However, existing agents struggle to achieve both decoupled enhancement and balanced integration of these...
MT-PingEval: Evaluating Multi-Turn Collaboration with Private Information Games
arXiv:2602.24188v1 Announce Type: new Abstract: We present a scalable methodology for evaluating language models in multi-turn interactions, using a suite of collaborative games that require effective communication about private information. This enables an interactive scaling analysis, in which a fixed...
Controllable Reasoning Models Are Private Thinkers
arXiv:2602.24210v1 Announce Type: new Abstract: AI agents powered by reasoning models require access to sensitive user data. However, their reasoning traces are difficult to control, which can result in the unintended leakage of private information to external parties. We propose...
Do LLMs Benefit From Their Own Words?
arXiv:2602.24287v1 Announce Type: new Abstract: Multi-turn interactions with large language models typically retain the assistant's own past responses in the conversation history. In this work, we revisit this design choice by asking whether large language models benefit from conditioning on...
Serendipity with Generative AI: Repurposing knowledge components during polycrisis with a Viable Systems Model approach
arXiv:2602.23365v1 Announce Type: cross Abstract: Organisations face polycrisis uncertainty yet overlook embedded knowledge. We show how generative AI can operate as a serendipity engine and knowledge transducer to discover, classify and mobilise reusable components (models, frameworks, patterns) from existing documents....
HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit
arXiv:2602.23699v1 Announce Type: cross Abstract: The quadratic computational cost of processing vision tokens in Multimodal Large Language Models (MLLMs) hinders their widespread adoption. While progressive vision token pruning offers a promising solution, current methods misinterpret shallow layer functions and use...
U-CAN: Utility-Aware Contrastive Attenuation for Efficient Unlearning in Generative Recommendation
arXiv:2602.23400v1 Announce Type: new Abstract: Generative Recommendation (GenRec) typically leverages Large Language Models (LLMs) to redefine personalization as an instruction-driven sequence generation task. However, fine-tuning on user logs inadvertently encodes sensitive attributes into model parameters, raising critical privacy concerns. Existing...
Active Value Querying to Minimize Additive Error in Subadditive Set Function Learning
arXiv:2602.23529v1 Announce Type: new Abstract: Subadditive set functions play a pivotal role in computational economics (especially in combinatorial auctions), combinatorial optimization or artificial intelligence applications such as interpretable machine learning. However, specifying a set function requires assigning values to an...
FlexGuard: Continuous Risk Scoring for Strictness-Adaptive LLM Content Moderation
arXiv:2602.23636v1 Announce Type: new Abstract: Ensuring the safety of LLM-generated content is essential for real-world deployment. Most existing guardrail models formulate moderation as a fixed binary classification task, implicitly assuming a fixed definition of harmfulness. In practice, enforcement strictness -...
Provable Subspace Identification of Nonlinear Multi-view CCA
arXiv:2602.23785v1 Announce Type: new Abstract: We investigate the identifiability of nonlinear Canonical Correlation Analysis (CCA) in a multi-view setup, where each view is generated by an unknown nonlinear map applied to a linear mixture of shared latents and view-private noise....
GRAIL: Post-hoc Compensation by Linear Reconstruction for Compressed Networks
arXiv:2602.23795v1 Announce Type: new Abstract: Structured deep model compression methods are hardware-friendly and substantially reduce memory and inference costs. However, under aggressive compression, the resulting accuracy degradation often necessitates post-compression finetuning, which can be impractical due to missing labeled data...
MPU: Towards Secure and Privacy-Preserving Knowledge Unlearning for Large Language Models
arXiv:2602.23798v1 Announce Type: new Abstract: Machine unlearning for large language models often faces a privacy dilemma in which strict constraints prohibit sharing either the server's parameters or the client's forget set. To address this dual non-disclosure constraint, we propose MPU,...
Learning to maintain safety through expert demonstrations in settings with unknown constraints: A Q-learning perspective
arXiv:2602.23816v1 Announce Type: new Abstract: Given a set of trajectories demonstrating the execution of a task safely in a constrained MDP with observable rewards but with unknown constraints and non-observable costs, we aim to find a policy that maximizes the...
Intrinsic Lorentz Neural Network
arXiv:2602.23981v1 Announce Type: new Abstract: Real-world data frequently exhibit latent hierarchical structures, which can be naturally represented by hyperbolic geometry. Although recent hyperbolic neural networks have demonstrated promising results, many existing architectures remain partially intrinsic, mixing Euclidean operations with hyperbolic...
pathsig: A GPU-Accelerated Library for Truncated and Projected Path Signatures
arXiv:2602.24066v1 Announce Type: new Abstract: Path signatures provide a rich representation of sequential data, with strong theoretical guarantees and good performance in a variety of machine-learning tasks. While signatures have progressed from fixed feature extractors to trainable components of machine-learning...
Inner Speech as Behavior Guides: Steerable Imitation of Diverse Behaviors for Human-AI coordination
arXiv:2602.20517v1 Announce Type: new Abstract: Effective human-AI coordination requires artificial agents capable of exhibiting and responding to human-like behaviors while adapting to changing contexts. Imitation learning has emerged as one of the prominent approaches to build such agents by training...
From Logs to Language: Learning Optimal Verbalization for LLM-Based Recommendation in Production
arXiv:2602.20558v1 Announce Type: new Abstract: Large language models (LLMs) are promising backbones for generative recommender systems, yet a key challenge remains underexplored: verbalization, i.e., converting structured user interaction logs into effective natural language inputs. Existing methods rely on rigid templates...
When can we trust untrusted monitoring? A safety case sketch across collusion strategies
arXiv:2602.20628v1 Announce Type: new Abstract: AIs are increasingly being deployed with greater autonomy and capabilities, which increases the risk that a misaligned AI may be able to cause catastrophic harm. Untrusted monitoring -- using one untrusted model to oversee another...
Recursive Belief Vision Language Model
arXiv:2602.20659v1 Announce Type: new Abstract: Current vision-language-action (VLA) models struggle with long-horizon manipulation under partial observability. Most existing approaches remain observation-driven, relying on short context windows or repeated queries to vision-language models (VLMs). This leads to loss of task progress,...
LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification
arXiv:2602.21044v1 Announce Type: new Abstract: Evaluations of large language models (LLMs) primarily emphasize convergent logical reasoning, where success is defined by producing a single correct proof. However, many real-world reasoning problems admit multiple valid derivations, requiring models to explore diverse...
Motivation is Something You Need
arXiv:2602.21064v1 Announce Type: new Abstract: This work introduces a novel training paradigm that draws from affective neuroscience. Inspired by the interplay of emotions and cognition in the human brain and more specifically the SEEKING motivational state, we design a dual-model...
ShaRP: Shape-Regularized Multidimensional Projections
arXiv:2306.00554v1 Announce Type: cross Abstract: Projections, or dimensionality reduction methods, are techniques of choice for the visual exploration of high-dimensional data. Many such techniques exist, each one of them having a distinct visual signature - i.e., a recognizable way to...
Interpretable Medical Image Classification using Prototype Learning and Privileged Information
arXiv:2310.15741v1 Announce Type: cross Abstract: Interpretability is often an essential requirement in medical imaging. Advanced deep learning methods are required to address this need for explainability and high performance. In this work, we investigate whether additional information available during the...
From Performance to Purpose: A Sociotechnical Taxonomy for Evaluating Large Language Model Utility
arXiv:2602.20513v1 Announce Type: new Abstract: As large language models (LLMs) continue to improve at completing discrete tasks, they are being integrated into increasingly complex and diverse real-world systems. However, task-level success alone does not establish a model's fit for use...
fEDM+: A Risk-Based Fuzzy Ethical Decision Making Framework with Principle-Level Explainability and Pluralistic Validation
arXiv:2602.21746v1 Announce Type: new Abstract: In a previous work, we introduced the fuzzy Ethical Decision-Making framework (fEDM), a risk-based ethical reasoning architecture grounded in fuzzy logic. The original model combined a fuzzy Ethical Risk Assessment module (fERA) with ethical decision...
ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices
arXiv:2602.21858v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have made significant progress in mobile agent development, yet their capabilities are predominantly confined to a reactive paradigm, where they merely execute explicit user commands. The emerging paradigm of proactive...