The Surprising Effectiveness of Deleting Weights in LLM Post-Training
In shortZeroing less than 0.05% of LLM weights via a learned binary mask matches full fine‑tuning across several post‑training tasks.
Published 6 Nov 2026
13 Nov 2026
200 papers that day.
Picks that day
Few-Step Streaming Video Generation via Condition Shifting
In shortAPEX removes the fake-score network for few-step streaming video generation by shifting conditions within the generator.
Published 10 Nov 2026
Everything else
198 more papers from this day-
Nov 10
Don't Just Trust Your Memory: Evidence-Linked Memory Reduces Memory-Induced Degradation in LLM Agents
TrendingReasoning & Agents
- Nov 10 MMDM: Measuring Massive Multitask Decision Making
- Nov 7 One Table, Many Queries: Benchmarking Generalist Tabular Models Across Generation, Imputation, and Prediction
- Nov 10 Tree-Ising: Repairing Language Models with Bit-Level Ising Edits
- Nov 9 Embodied First Exam
- Nov 6 MACaQ: Fast, Model Agnostic Calibrated Quantization for Low-Precision Weights and Activations
- Nov 9 GTAGym: Real AAA Game Tasks Environment for Evaluation and Reinforcement Learning
- Nov 8 Evolving Infra: Agent-Adaptive LLM Serving at Scale
- Nov 4 Interactive Visual Gym: Evaluating Visual Interaction Models with Simulated Users in Interactive Worlds
- Nov 9 SteerDistill: Quantizing LLM Agents via Closed-Loop Teacher Steering
- Nov 9 WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning
- Oct 28 Scaling Mathematical Proof Supervision from Structured Textbooks
- Oct 28 DyCraft: An Agentic System for Long-Horizon 4D Scene Generation
- Nov 9 Learning to Act with World Models by Valuing What Comes Next
- Nov 9 Blind Variation, Selective Retention: Improving Creative Idea Generation in AI Scientists
- Nov 9 Where Does Alignment Emerge and What Does It Mean? A Closer Look at the Platonic Representation Hypothesis
- Oct 28 Incentivizing Self-Learning in Large Language Models via Reinforcement Learning
- Nov 10 LEASH: Stay Safe in Agentic Robotics
- Nov 9 FAME-QP: A GPU-Accelerated Quadratic-Programming Solver for Real-Time Model Predictive Control
- Nov 10 PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video Generation
- Nov 10 Market-Conditioned JEPA: Learning from Unpredictable Market Shocks
- Nov 5 EarthRecord: From Pixels to Land Change with Multimodal Agents over Four Decades of Satellite Imagery
- Nov 4 FINCOSMOS: An Open-World Evaluation Paradigm for Agents in Finance
- Nov 10 Sable: A Small-Field Sumcheck Argument for Bit-Exact Inference of Large Language Models
- Nov 10 Are Causal Neuron Circuits Unique and Transferable?
- Nov 6 Read Less, Solve More: Token-Efficient Sparse Reading of External Content for AI Agents
- Oct 29 CogBattery: Human-Referenced Evaluation of Cognitive Task Behavior and Structure in Language Models
- Nov 10 TorchCurves: Differentiable Parametric Curves
- Nov 10 EVAThink: Evidence Sufficiency-Guided Training for Adaptive Thinking in Video Understanding
- Nov 10 Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model
- Nov 3 CARTBench: Rubric-Based Evaluation of Shopping Advice for Complex and Ambiguous Requests with Timelines
- Nov 4 Module-Aware Optimization for Graph Neural Networks
- Oct 29 WMIP: A Working-Memory-Inspired Programmable Probe of Cumulative State Tracking in Large Language Models
- Nov 10 ZoomGain: Mining Global-to-Zoom Perception Gaps for Fine-Grained Visual Supervision
- Nov 9 ScrapeBench: Evaluating Legal Compliance of AI Agents in Website Scraping
- Oct 31 From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research Agents
- Nov 5 Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference
- Nov 6 Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking
- Nov 3 Micius-Flash: Lightweight Agentic Intelligence for Engineering
- Nov 10 Visual Similarity in Adult Word Extension and Artificial Word Learning
- Nov 9 Mixed Signals: Weight-Space Programming for EEG and ECG
- Nov 9 LO-World: Unified BEV Latent Modeling for Joint 4D Occupancy and LiDAR Forecasting
- Nov 10 Are Dense Rewards Worth Their Tokens in GRPO?
- Nov 9 Shortcuts and Boundaries: Verifiable RL Misalignment and Contextual Safety Learning
- Nov 10 RS-JEPA: Remote Sensing Joint-Embedding Predictive Architecture with Scan-Path Views for Cross-Sensor Image Retrieval
- Nov 10 REAL Loss: Optimizing Real Expected Acceptance Length for Speculative Decoding
- Nov 10 From Video Frames to Robot Meta-Frames: A Unified Latent Interface for World Action Models
- Nov 10 Forward-Value Discretization: Faster Stochastic Diffusion Sampling Without High-Order Approximation
- Nov 10 Multimodal Abstention That Preserves Acquired Correct Answers
- Nov 10 TaoMate-R1: Temporal Role Specialization for Streaming Multi-Reference Video Generation
- Nov 10 Location Matters: Sparse Latent-Variance Attacks for Face Privacy against Diffusion Personalization
- Nov 10 HaiRuo: Forecasting Ocean Eddies as Fields and Objects with a Satellite-Altimetry Foundation Model
- Nov 10 Sim(3)-TurboReg: Robust and Efficient Scale-Aware Registration via Tri-Scale Consistency
- Nov 10 Hierarchical Graph Representation Learning with Pooling-Induced Substructures
- Nov 10 HiPFuse: Real-Time LiDAR Semantic Segmentation for Robotic Perception on Edge Devices
- Nov 10 LUMIR: Decoupled Multimodal Representation Learning from Heterogeneous Data for Next-POI Recommendation
- Nov 10 Specify Privacy Yourself: Assessing Inference-Time Personalized Privacy Preservation Ability of Large Vision-Language Models
- Nov 10 Understanding Mixture-of-Experts Language Models Beyond Expert Boundaries
- Nov 10 MicroBT: Centroid-Guided Synergistic Learning for Multi-Modal Micro Brain Tumor Segmentation and Counting
- Nov 10 Region-Wise Logit Fusion and Utility-Gated Refinement for Brain Metastasis Segmentation
- Nov 10 Towards Text-Only Remote Sensing Scene Classification via Semantic Proxy
- Nov 10 Bridging Graph and Pairwise Learning for Gene Regulatory Network Inference from scRNA-seq Data
- Nov 10 DisasterAgent: An Evidence-Driven Agent for Verifiable Disaster Report Generation From Remote Sensing Imagery
- Nov 10 Deterministic, Reproducible Inference in Exact Integer Arithmetic, and Verifiable Distributed Inference as Its Consequence
- Nov 10 AutoManifold: Agentic Design of Data Visualisation Algorithms via Manifold Embedding
- Nov 10 OpticalRAG: Pixel-Space Compression for Token-Efficient Retrieval-Augmented Generation
- Nov 10 CP-MLPs: A Tensor-Rank Theory of Tied and Untied MLP Blocks
- Nov 10 FlashMotion: Ultra-Fast Long-Term 3D Motion Generation
- Nov 10 Seeing the Small: Deformable Duality Mamba for Retinal Lesion Segmentation
- Nov 10 TMAL: Topology-preserving Multi-offset Affinity Learning for Curvilinear Structure Segmentation
- Nov 10 Mobile AI Factory: Building Specialized Language Models for Your Mobile Device
- Nov 10 Camera3R: Feed-Forward 3D Reconstruction from Heterogeneous Camera Models
- Nov 10 How Hyperbolic Are Hyperbolic Vision-Language Models?
- Nov 9 MRFoundry: Knowledge-guided agent loops for end-to-end MRI research
- Nov 9 Multimodal Judges Flag, Humans Decide: A Review-and-Edit Framework for AI-Generated Product Video Ads
- Nov 9 How Much of Your VLA Is kNN
- Nov 9 VIGOR: Benchmarking Visual Rationale Correctness in Multimodal Large Language Models
- Nov 9 Guiding Data Allocation for Robust Subpopulation Generalization
- Nov 9 Beyond Modality Preference: Controlling Multimodal Arbitration under Modality Conflict in MLLMs
- Nov 9 Mitigating Agentic Deception via Evaluation-Side Partially Observable Games
- Nov 9 Anchor Behavior, Not Parameters: Rehearsal-Free Continual Learning with a Single Shared Prompt
- Nov 9 PORT: Preference Optimization via Robust Token-Level Reweighting
- Nov 9 The Indic Token Tax in 2026: Measuring Tokenizer. Disparity and Its Cost for the Languages of India
- Nov 9 A four-step validation protocol for radiomic survival analysis in small cohorts: a body composition case study
- Nov 9 What Breaks Scribble-Supervised Segmentation? An Audit-Informed Benchmark for Scribble Annotation Robustness
- Nov 9 When Does Sliced Inference Pay Off? Resolution, Tiling and Dense Fusion for Small Object Detection
- Nov 9 Learning inexact alternating minimization for convex problems with quadratic coupling
- Nov 9 MARS: Multiscale Appearance-Residual Streams for Video Forgery Detection
- Nov 9 Perturb, Aggregate then Reconstruct: Consensus Aggregation and Fourier-Attended Multi-Decoders for Reconstruction-based Multi-View Anomaly Detection
- Nov 9 MoSTE: Mixture of Spatio-Temporal Experts for Deepfake Detection in Videos
- Nov 9 Patching Up Circuit Learning via Parameter Sharing
- Nov 9 More Than Just Style: Benchmarking Personality Influence in Large Language Models
- Nov 9 AutoVLA-D: Learn a Trajectory-Geometric Condition from Reasoning for Autonomous Driving VLA
- Nov 9 CADMA: Capacity-Aware Coverage Decomposition for Generative Model Assessment
- Nov 9 A Geometric Invariant of Attention Predicts Head Specialization from Weights Alone
- Nov 9 GeoMind: A Benchmark for Spatial Understanding in Robotic Manipulation
- Nov 9 One-Shot Federated Graph Learning via High-Fidelity Proxies and Transferability-Guided Collaboration
- Nov 9 Zero-Search Reinforcement Learning for Slay the Spire 2 with Failure-Driven Restarts
- Nov 9 Weight-Space Programming for Medical Imaging: Do We Still Need Training Data?
- Nov 9 AscendMOEA: An Ascend NPU-Accelerated Platform for Tensorized Multi-Objective Evolutionary Algorithms
- Nov 9 FocusBranch: Combinatorial Branch-and-Bound for $\ell_0$ Neural Network Robustness Verification
- Nov 9 HyPLM Technical Report
- Nov 9 HyInv Technical Report
- Nov 9 MUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice Protection
- Nov 9 AnyBreak: Towards Jailbreaking Any T2I Model for Any Concept
- Nov 9 3R-Adapter: Retrieval, Rewiring, and Refinement for Efficient Adaptation of 3D Reconstruction Model
- Nov 9 UltraVoxelGS: Voxel-First Feed-Forward Gaussian Splatting for 3D Ultrasound Reconstruction
- Nov 9 Can LLM Agents Organize Their Own Memory?
- Nov 9 LBSNet: Lightweight Brain State Modeling via Learnable Spectral Prototypes for EEG-Based Fatigue Detection
- Nov 9 Spectral Unlearning: Transformer Structure-Preserving Updates for Language Models
- Nov 9 Adaptive Entropy-Sparing for Efficient Reasoning
- Nov 9 Answer or Defer? Testing VLMs' Stated Confidence on Brain MRI
- Nov 9 When to Trust Adaptation: Contextual Fusion for Transferable Vision-Language Prompting
- Nov 9 SpinSCI: Mask-Conditioned Spatial-Temporal Spin Transformer for Video Snapshot Compressive Imaging
- Nov 9 Breaking the $\sqrt{d}$ Communication Barrier in Federated Sampling with Adaptive Hamiltonian Monte Carlo
- Nov 3 Subliminal Learning in Vision-Language Models
- Nov 7 Probing Transformer Representations Layer by Layer, Module by Module
- Nov 9 Precision on Demand: Escalating Weight Bits at Uncertain Decoding Steps in a Small Local Model
- Nov 10 Lipschitz-Controlled Flow Matching for Robust Image Restoration
- Nov 10 One Policy for 100 Household Tasks: Our Solution for the 2026 BEHAVIOR Challenge
- Nov 10 Earlybird: Predictive Streaming Speech Recognition in 29M Parameters
- Nov 9 Bits or permutations? Spending a fixed inference budget on weight precision or on option-order averaging in small local language models
- Nov 10 BioXArena-Pro: a research-grade benchmark for autonomous biomedical machine learning agents
- Nov 10 Alania-2: Turkish text-to-speech with instant and fine-tuned voice cloning
- Nov 9 PolyRoute: A Twelve-Language and Code-Mixed Benchmark for Typed-Decision Routing, with Baselines and a Leaderboard
- Nov 10 Antalia-2 Mini: A 7.6-million-parameter Turkish text-to-speech model trained entirely on synthetic speech
- Nov 10 Camera equalisation of an 11.5x drop in environmental illuminance: four imitation-learning policies on an SO-101 arm
- Nov 10 RF3DGS: Reflection-Free 3D Gaussian Splatting with Self-Priors
- Nov 10 One Byte, One Token: Byteifying a 27B Hybrid-Attention LLM with Self-Generated Data
- Nov 10 Towards Reliable VLM Judges: State-Conditional Invariance and Presentation-Aware Diagnostics
- Nov 10 Massive Activations Confound Sparse Autoencoders
- Nov 9 ExpertCrowd-SWE: Measuring Frontier Coding Agents on Expert-Authored Engineering Tasks in Unfamiliar Codebases
- Nov 9 dNaViT: Discrete Codes are General-Purpose Visual Representations
- Nov 9 How a Model's Architecture and Performance Influences Explainability: A Study of Grad-CAM and CRAFT in CNN-Based Image Classification
- Nov 10 Medical Image Segmentation: Foundation Model Adaptation versus Task-Specific Training
- Nov 10 Toward Foundation Detection Models via Example-Based Grounding in EM-DETR
- Nov 10 AutoScaffold: AI-Guided Scaffolding for Reinforcement Learning of Language Models
- Nov 10 Contrastive Discovery: Open-Ended Scientific Discovery over Competing Explanations
- Nov 10 Pyrite: Building and Benchmarking Protein-Ligand Pose Ranking Agents
- Nov 10 When Does the Real Wiring Win? Testing connectome structure-function claims against unmeasured physiology
- Nov 10 Continual-ARC: Measuring Skill Acquisition over a Working Lifetime
- Nov 10 Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
- Nov 10 Measuring Robustness and Efficiency in a Connectome-Constrained Fly Visual System Model: Linking an Adversarial and a Perceptual Illusion
- Nov 10 Comparative Analysis of Agent Evaluation Frameworks: Stability, Detection, and Discovery in Enterprise Analytics Agents
- Nov 10 ReTrace: Distilling Tool-Using Agents into Planner-Free Workflows
- Nov 10 TimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series Analytics
- Nov 10 Is Your Robot's Memory Just a Clock? Temporal Shortcuts in Memory-Augmented VLA Policies
- Nov 10 Re-Search: Dissecting LLM Ideation Through a Two-Player Questioning Game
- Nov 10 Mechanistic Interpretability Needs a Logic of Inference
- Nov 10 Pterodactyl: Pushing the Limits of Efficient Automated Translation Evaluation
- Nov 10 FERN: Cross-Benchmark Table Retrieval for Text-to-SQL with Frozen Retrievers and Learned Ranking
- Nov 10 BoviShift: A Six-Year Multi-Attribute Audit of Visual Cattle Phenotyping Across Animals and Time
- Nov 10 TanCRAFT: Context-Reasoned Anticipatory Feedback for Assembly Tasks
- Nov 10 C$^2$-Routing: Complementary Cropping and Compression for Efficient Vision-Language Inference
- Nov 10 Bayesian inverse problems and epidemic-threshold uncertainty for a heterogeneous SIS reaction-diffusion model
- Nov 10 A Theory of Policy Optimization in Post-Training: Coverage Collapse and Reference Resets
- Nov 10 One Curve, Different Weights: AUC, the Brier Score, Decision-Weighted AUC, and the Policy Youden Index
- Nov 10 A Theory of Cross-Validation Variance
- Nov 10 Accurate predictions of properties of energetic materials with tabular foundation models
- Nov 10 When does text encoding matter for language modeling?
- Nov 10 A Lightweight Differentiable Method for Real-Time Skin Rendering and Digitization
- Nov 10 Nonparametric Maximum Likelihood Estimation for Mixtures of Logistic Regressions
- Nov 10 UGG: Unified Generative Geometry Model
- Nov 10 Metric-Aligned Group Relative Policy Optimization for Explainable Deepfake Detection
- Nov 10 Two Communities, One Problem: Branch-and-Bound vs. A* for Resource-Constrained Project Scheduling
- Nov 10 Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications
- Nov 10 DnNP: Denoising Input Uncertainty in Neural Processes
- Nov 10 OnlyAct: Extracting Action Representations from Foundation Models for World Models
- Nov 10 Evaluating LLM-Generated World Models for Neuro-Symbolic Reasoning
- Nov 10 Exploring Post-Hoc Disentanglement of Audio Foundation Models for Music Similarity
- Nov 10 SIGMO: Sigmoid Gated Interaction-Aware Motion Planning
- Nov 10 Causal Concept Explanations for Deep Neural Models
- Nov 10 Discovering Equations for the Thermodynamics of South Asian Monsoon Precipitation
- Nov 10 A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA)
- Nov 10 Blurring the Block Boundaries: Overlapping Block Denoising for Efficient Diffusion Language Models
- Nov 10 An Energy-Based Perspective on Machine Learning
- Nov 10 Cognitive Mapping of Learning by Thinking Mechanisms in Large Language Models
- Nov 10 What Factors Matter for Collection-Level Counting with Human-AI Collaboration?
- Nov 10 Uncertainty-Aware Hybrid Offline Reinforcement Learning with Online Action Selection Enabled by Sparse Gaussian-Process Critics
- Nov 10 Gaussian Process-Augmented Model Predictive Contouring Control for 6-DoF Trajectory Tracking with a Holonomic AUV
- Nov 10 ARE-bench: Benchmarking Agents' Research Ability in Real-World Researcher Sessions and Workspaces
- Nov 10 The Unused Input: The Initial State as a Conditioning Interface for Recurrent TSFMs
- Nov 10 EngineAgent: Agentizing On-Device LLM Inference Engines with Inference Stack Hierarchy Awareness and Hardware Abilities
- Nov 10 CLADE: Category-Level Answering from Deficient Evidence in Retrieval-Augmented Language Models
- Nov 10 AI Isn't Getting Dumber, We're Getting Pickier: Rising Standards in Human Evaluation of Large Language Models
- Nov 10 When Should Agents Remember? Falsification-Gated Self-Evolution for LLM Agents
- Nov 10 Point-in-Time Encoders for Earnings Calls
- Nov 10 Auditing the Judge: Internal Signals for LLM Reasoning Assurance
- Nov 10 MammothDP: Private Boosted Decision Trees Ready for Trusted Hardware and Leakage-free Hyperparameters
- Nov 10 PSL-RGHA: Parallel Split Learning with Reference Gradient Half-Space Alignment for Mitigating Gradient Divergence
- Nov 10 ForeseeDrive: Future-Token Conditioning for Anticipatory Generative Driving
- Nov 10 LM-SimBench: A Unified Framework, Dataset, and Benchmark for Large Language Model Training Simulation
- Nov 10 Ninety-Six Ways to Measure a Scribe: Fragmented Evaluation of Ambient Scribes in Healthcare
- Nov 10 UniFlow-RS: Unified Generative Prediction for Heterogeneous Remote-Sensing Tasks
- Nov 10 AutoThreshold: Joint Confusion-Profile Learning for Adaptive Mask Binarization
- Nov 10 Compositional random forest
- Nov 10 From Prediction to Action: Learning Physical Context for Deformable Linear Object Manipulation
- Nov 10 Federated Learning for Cross-Regional Epidemic Forecasting with Asynchronous Outbreaks