Search: “models”

MAAT: Multi-phase Adapter-Aware Targeted Unlearning

model-releasesarxiv-cs-cl

arXiv:2605.30514v1 Announce Type: cross Abstract: Machine unlearning evaluation is structurally skewed: Why-type questions, which probe causal and relational knowledge, comprise less than 0.06% of Cou

Applications

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

MechVQA: Benchmarking and Enhancing Multimodal LLMs on Comprehensive Mechanical Drawing Understanding

DGX agent

arXiv:2605.30794v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they re

applicationsarxiv-cs-ai

1 Jun 2026

Local Ai

Memory-Efficient Structured Backpropagation for On-Device LLM Fine-Tuning

arXiv:2602.13069v2 Announce Type: replace-cross Abstract: On-device fine-tuning enables privacy-preserving personalization of large language models, but mobile devices impose severe memory constraints

local-aiarxiv-cs-cl

Mixture of Concept Bottleneck Experts

arXiv:2602.02886v2 Announce Type: replace-cross Abstract: Concept Bottleneck Models (CBMs) promote interpretability by grounding predictions in human-understandable concepts. However, existing CBMs ty

Local Ai

Mixture of Horizons in Action Chunking

arXiv:2511.19433v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the extb

local-aiarxiv-cs-ai

Neuro-symbolic Syntactic Parsing: Shaping a Neural Network with the CYK Algorithm

arXiv:2605.31421v1 Announce Type: cross Abstract: In this paper, we show the possibility of a direct injection of algorithms into neural network architecture. We focus on a complex algorithm, that is,

nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving

arXiv:2605.31572v1 Announce Type: new Abstract: Reasoning is essential for autonomous driving (AD) in long-tail scenarios, where vehicles must apply commonsense knowledge, understand spatial relations

On the impact of retrieved content representations in RAG Pipelines

arXiv:2605.30790v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) supplements a language model's input with retrieved documents, yet most RAG pipelines inherit retrieval component

On the Relationship Between Activation Outliers and Feature Death in Sparse Autoencoders

arXiv:2605.31518v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) decompose neural network activations into interpretable features, but many learned features never activate, a problem called

safetyarxiv-cs-lg

Parameter-free Dynamic Regret: Time-varying Movement Costs, Delayed Feedback, and Memory

arXiv:2602.06902v2 Announce Type: replace Abstract: In this paper, we study dynamic regret in unconstrained online convex optimization (OCO) with movement costs. Specifically, we generalize the standa

Physics Enhanced Deep Surrogates for the Phonon Boltzmann Transport Equation

arXiv:2512.05976v3 Announce Type: replace-cross Abstract: Designing materials with controlled heat flow at the nano-scale is central to advances in microelectronics, thermoelectrics, and energy-conver

Applications

Positional versus Symbolic Attention Heads: Learning Dynamics, RoPE Geometry, and Length Generalization

arXiv:2605.31558v1 Announce Type: cross Abstract: Transformer-based language models are widespread in today's society. As such, understanding the mechanisms by which they solve structured tasks and pr

applicationsarxiv-cs-ai

Probing the Prompt KV Cache: Where It Becomes Dispensable

model-releasesarxiv-cs-cl

arXiv:2605.30574v1 Announce Type: new Abstract: Prior KV cache compression schemes empirically demonstrate that the prompt cache is partially redundant during decoding, dropping or summarising entries

Randomized Feasibility Methods for Constrained Optimization with Adaptive Step Sizes

arXiv:2601.20076v2 Announce Type: replace-cross Abstract: We consider minimizing an objective function subject to constraints defined by the intersection of lower-level sets of convex functions. We st

Rationalize: Shared Semantic Reasoning for Human-AI Alignment

arXiv:2605.30632v1 Announce Type: cross Abstract: We introduce Rationalize, a role-pair framework for shared semantic reasoning between humans and AI models in data-driven sensemaking. Building on ide

Tutorials

Reliable Self-Improvement Training by Verifying Reasoning, Not Just Answers

arXiv:2603.21558v2 Announce Type: replace Abstract: Self-improvement training, where models learn from self-generated solutions, promises sustained capability gains but suffers from a pervasive failur

tutorialsarxiv-cs-ai

Scalable Inference-Time Annealing with Surrogate Likelihood Estimators

arXiv:2605.31498v1 Announce Type: new Abstract: A long standing challenge in computational chemistry and biophysics is efficiently sampling the Boltzmann distribution of molecules. Advances in generat

researcharxiv-cs-lg

Self-Reflective Generation at Test Time

arXiv:2510.02919v2 Announce Type: replace Abstract: Large language models (LLMs) increasingly solve complex reasoning tasks via long chain-of-thought, but their forward-only autoregressive generation

researcharxiv-cs-cl

Self-Tuning Regularization for Image Scanning Microscopy

arXiv:2605.31426v1 Announce Type: cross Abstract: Image Scanning Microscopy (ISM) is a fluorescence imaging technique that combines detector-array acquisition and computational reconstruction to achie

Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense

arXiv:2605.30837v1 Announce Type: cross Abstract: Prompt-injection detectors are heterogeneous: each is strong on a different slice of attacks, and none is always reliable. Yet existing systems still

SimulCost: A Cost-Aware Benchmark and Toolkit for Automating Physics Simulations with LLMs

arXiv:2603.20253v2 Announce Type: replace-cross Abstract: Evaluating LLM agents for scientific tasks has focused on token costs while ignoring tool-use costs like simulation time and experimental reso

SLAT: Segment-Level Adaptive Trimming for Efficient CoT Reasoning

arXiv:2605.30832v1 Announce Type: new Abstract: Recent advances in Large Reasoning Models have significantly improved chain-of-thought (CoT) capabilities via reinforcement learning (RL). However, gene

Social welfare optimisation under institutional reward and punishment

arXiv:2605.31330v1 Announce Type: cross Abstract: Institutional incentives are widely used to promote cooperation among autonomous, self-regarding agents, from human societies to multi-agent and AI sy

Softsign: Smooth Sign in Your Optimizer For Better Parameter Heterogeneity Handling

arXiv:2605.31371v1 Announce Type: new Abstract: Sign-based and LMO-inspired optimizers have recently attracted substantial attention in deep learning due to their strong performance and low memory foo

Stateful Online Monitoring Catches Distributed Agent Attacks

arXiv:2605.31593v1 Announce Type: cross Abstract: Language models can find thousands of severe software vulnerabilities, and agents are increasingly being misused for cyberattacks. To avoid detection,

The Geometry of Activity Cliffs: Representation Dependence and Multi-Scale Characterization of Activity Landscapes

arXiv:2605.30831v1 Announce Type: cross Abstract: Activity cliffs, structurally similar compounds with large potency differences, are widely treated as intrinsic features of chemical datasets. We argu

The Global Landscape of Environmental AI Regulation: From the Cost of Reasoning to a Right to Green AI

arXiv:2603.00068v2 Announce Type: replace-cross Abstract: Artificial intelligence (AI) systems impose substantial and growing environmental costs, yet transparency about these impacts has declined eve

Towards Efficient LLMs Annealing with Principled Sample Selection

arXiv:2605.31175v1 Announce Type: new Abstract: The annealing phase is a pivotal convergence stage in LLM pre-training that ultimately determines final model quality. However, effectively selecting tr

researcharxiv-cs-cl

Toxic HallucinAItions: Perturbing Prompts and Tracing LLM Circuits

arXiv:2605.30913v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in conversational settings where user tone ranges from polite to adversarial or toxic, yet less

TUX: Measuring Human--AI Tacit Understanding

arXiv:2605.30930v1 Announce Type: cross Abstract: As large language models (LLMs) increasingly act as collaborative partners, human--AI alignment is often evaluated through explicit task success, accu

Unfolding Generative Flows with Koopman Operators: Trajectory-Preserving Linearization

arXiv:2506.22304v3 Announce Type: replace-cross Abstract: Continuous Normalizing Flows (CNFs) enable elegant generative modeling but remain bottlenecked by their iterative nature requiring costly samp

safetyarxiv-cs-cv

Variational Adapter for Cross-modal Similarity Representation

arXiv:2605.30968v1 Announce Type: cross Abstract: The core of vision-language models lies in measuring cross-modal similarity within a unified representation space. However, most image-text matching o

Applications

VisionPulse: Dynamic Visual Sparsity for Efficient Multimodal Reasoning

arXiv:2605.31457v1 Announce Type: new Abstract: With the rapid advancement of large multimodal models (LMMs), inference-time overhead has become a key bottleneck for real-world deployment. Existing me

applicationsarxiv-cs-cv

Local Ai

What changes after deployment? A survey on On-device Learning in TinyML

arXiv:2605.31226v1 Announce Type: cross Abstract: Machine learning models on microcontroller-class devices (TinyML) face a fundamental challenge: post-deployment distribution change undermines static

local-aiarxiv-cs-ai

A Computational Method for Solving the Stochastic Joint Replenishment Problem in High Dimensions

arXiv:2511.11830v2 Announce Type: replace-cross Abstract: We consider a discrete-time formulation for a class of high-dimensional stochastic joint replenishment problems. First, we approximate the pro

A new completely parameter-free clustering algorithm for unsupervised classification of BATSE gamma-ray bursts

arXiv:2605.30175v1 Announce Type: cross Abstract: Cluster analysis is a widely applied machine learning technique to understand the existing patterns in the population of gamma-ray bursts (GRBs), in o

A Progress-Aware Leader-Follower Midair Docking System for Dual-Drone Aerial Manipulation

model-releasesarxiv-cs-ro

arXiv:2605.29410v1 Announce Type: new Abstract: Reliable midair docking between small unmanned aerial vehicles (UAVs) is essential for modular aerial cooperation and manipulation, but it requires prec

Adaptive Interviewing for Persona Simulation in LLMs: Evidence-Grounded Reasoning Improves Decision Alignment

arXiv:2605.29458v1 Announce Type: cross Abstract: Accurately simulating the decisions of a specific individual remains challenging for large language models (LLMs), partly because persona information

AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning

arXiv:2602.23258v2 Announce Type: replace Abstract: While Multi-Agent Systems (MAS) excel in complex reasoning, they suffer from the cascading impact of erroneous information from individual agents. C

An accuracy-aware extension to LRP-based pruning for CNNs to prevent cascading accuracy degradation in data-scarce transfer learning

arXiv:2511.10861v3 Announce Type: replace-cross Abstract: Convolutional Neural Networks (CNNs) pre-trained on large-scale datasets such as ImageNet are widely used as feature extractors to construct h

Audio Deepfake Detection with Half-Truth Localisation Using Cross-Attentive Feature Fusion

arXiv:2605.29531v1 Announce Type: cross Abstract: Audio deepfake detection is well-studied as a binary problem, but partially manipulated speech, where a short synthesised segment is spliced into an o

Auditing Training-Free 3D Shape Retrieval with Diffused Geodesic Moments

arXiv:2605.29004v1 Announce Type: new Abstract: Reported retrieval scores for training-free shape descriptors conflate local signal design, normalization, aggregation, codebook fitting, and metric cho

BrahmicTokenizer-131K: An Indic-Capable Drop-In Replacement for o200k_base

model-releasesarxiv-cs-cl

arXiv:2605.29379v1 Announce Type: new Abstract: We present BrahmicTokenizer-131K, a 131,072-vocabulary byte-level BPE tokenizer that closes the Brahmic compression gap at the 131K-vocabulary class whi

CCS: Clinical Consensus Selection for Radiology Report Generation

arXiv:2605.30131v1 Announce Type: new Abstract: Radiology report generation (RRG) is commonly formulated as a single-path generation task, where a multimodal large language model (MLLM) produces one d

researcharxiv-cs-cl

CityGen: Structure-Guided City-Style Synthesis for Cross-City Autonomous Driving

arXiv:2605.29935v1 Announce Type: cross Abstract: Autonomous driving systems are commonly trained and evaluated within limited geographic regions, which hinders their scalability when deployed in new

Coarse-Grained Boltzmann Generators

arXiv:2602.10637v2 Announce Type: replace Abstract: Sampling equilibrium molecular configurations from the Boltzmann distribution is a longstanding challenge. Boltzmann Generators (BGs) address this b

researcharxiv-cs-lg

Colored Noise Diffusion Sampling

arXiv:2605.30332v1 Announce Type: new Abstract: Diffusion models achieve state-of-the-art image synthesis, with their generative trajectories fundamentally exhibiting a spectral bias, resolving low-fr

safetyarxiv-cs-cv

Compute Allocation in Evolutionary Search: From Depth-Breadth to Multi-Armed Bandits

arXiv:2605.29268v1 Announce Type: cross Abstract: LLM-guided evolutionary search (Evolve systems) has reached state-of-the-art results on mathematical and combinatorial tasks, yet most existing system