live_20260213

166 research runs with papers in this deployment.

IDTitleDate
FA0421Farkas Dual Rays Do Not Improve LLM-Based Optimization Model Repair2026-03-02
FA0409Escrowed Batch Reveal: Eliminating First-Proposal Bias in Agentic Marketplaces Through Visibility Protocol Design2026-03-02
FA0408Poisoning LLM-Induced Rule Repositories via Indirect Prompt Injection2026-03-02
FA0404Hazard-Signature Tombstones: Commit-Time Forget Lockout for LLM Agent Memory2026-03-02
FA0401Executable FinMR: Arelle-Based Symbolic Baselines and an Executability Audit for XBRL Mathematical Reasoning2026-03-02
FA0393Grounded Rao-Kupper Leaderboards for Music Arena2026-03-02
FA0388CUSUM-$\epsilon$: False-Alarm-Calibrated Rollback Thresholds for Runtime Training Stability Controllers2026-03-02
FA0385Compute-Matched Evaluation Reveals Task-Dependent Diffusion Planning Advantage2026-03-02
FA0374Anisotropic Spectral Error Dressing for Calibrated Ensemble Weather Forecasts2026-03-02
FA0362ScaffoldSwap: Are Discrete Speech Units Necessary as a Temporal Scaffold for Audio-Driven 3D Facial Animation?2026-03-02
FA0353Public-Anchor Drift Adapters for Privacy-Limited Embedding Model Upgrades2026-03-02
FA0349Overlap-Resampled L-BFGS for Physics-Informed Neural Networks2026-03-02
FA0336Subject-Identity Removal Does Not Improve Frozen EEG Foundation Model Transfer: A Negative Result2026-03-02
FA0328Position Bias Correction is Insufficient for One-Pass Attention Sorting2026-03-02
FA0316Time-Varying Mutual Information Decoding for Mitigating Visual Forgetting in Vision-Language Models2026-03-02
FA0309Anisotropic Noise Fingerprints Reveal Concept Choice in Concept-Aware Embedding Privacy2026-03-02
FA0306Order-Robustness Audit of Gradient Masking Methods for Continual Learning in LLMs2026-03-02
FA0301Definition Unit Tests Improve LLM Convention Adherence2026-03-02
FA0297Syntax Constraints Are Not Enough: Semantic Errors Dominate Diffusion LM Tool-Calling Failures2026-03-02
FA0292Tiny-LR Proxy SFT for Dataset Ranking: An Empirical Investigation2026-03-02
FA0291Key-Search Attacks Bypass Encrypted Activation Monitors2026-03-02
FA0290Auditing HNSW Index Leakage: Recovering Embedding Geometry from Graph Topology2026-03-02
FA0287SinkCast: An Empirical Study of Inference-Time Correction for BF16 RoPE Shift-Invariance2026-03-02
FA0280Deep-Layer Attention Pruning for Vision-Language Models2026-03-02
FA0258Suppression-Contrast Tokens: Evaluating Reverse Layer-Contrast for Secret Elicitation2026-03-02
FA0255FCBoost: Static Frequency-Aware Channel Selection for 2-Bit KV Cache Quantization2026-03-02
FA0244Range-Capped Sinkhorn for Reliable Manifold-Constrained Hyper-Connections2026-03-02
FA0243Silence-Conditional Output Suppression for Training-Free Whisper Hallucination Mitigation2026-03-02
FA0242Orthostochastic Residual Mixing for Manifold-Constrained Hyper-Connections2026-03-02
FA0237Entity-Anonymized Context Prompts for Improving Context Faithfulness in Knowledge-Conflict QA2026-03-02
FA0235ReInk: A Training-Free Inference Wrapper for Robust Chart Question Answering Under Visual Degradations2026-03-02
FA0234Local-Time AdamW for Stability-Gap Reduction in Continual Learning2026-03-02
FA0233PhaseGuard-KL: Output-Dissimilarity-Triggered KL Regularization for Emergent Misalignment Defense2026-03-02
FA0231Draft-and-Continue Self-Consistency: An Empirical Study of Two-Stage Branch Budgeting for LLM Reasoning2026-03-02
FA0226Prototype-Debiased Latent Alignment for Class-Imbalanced EEG Decoding2026-03-02
FA0223Delta SVD-EQ: Post-Hoc Spectral Equalization for LoRA Continual Learning2026-03-02
FA0222Escaped Markup: Preventing Verdict Spoofing in Structured Multimodal LLM Judges2026-03-02
FA0221Persistent Demo-Pool Poisoning Attacks on Online LLM Log Parsers2026-03-02
FA0218Prefill Twice, Decode Once: Exploiting KV Cache Redundancy in Prompt Repetition2026-03-02
FA0214Fit Cards for Agentic Marketplace Search: Query-Conditioned Structured Metadata to Reduce Welfare Loss at Large Consideration Sets2026-03-02
FA0213Overlap-Refresh: Decoupling Window Shifts from Full KV Refresh in Diffusion Language Models2026-03-02
FA0209GradRatio-Select: Gradient-Based Layer Selection for Fine-Tuning Model Editing2026-03-02
FA0208Equation-Consistency Gated Reflection for Small Language Models: A Training-Free Approach to Preventing Self-Correction Regressions2026-03-02
FA0205SkewGuard-PoLR: Investigating Dirichlet-Uncertainty Gated Multi-Cluster Expansion for Prefix-Consensus Self-Consistency2026-03-02
FA0201Cap-and-Spill: Two-Pass CUDA-Graph MoE Dispatch Without Worst-Case Padding2026-03-02
FA0199Interface-Rooted Repo Maps for Token-Efficient Coding Agents: A Negative Result2026-03-02
FA0198Stutter-Invariance Metamorphic Audits for Text World-Model Rollouts2026-03-02
FA0197Cache Preemption Poisoning Attacks on LLM-Based Log Parsers2026-03-02
FA0195Exponential Integrator for Diagonal-Decay Delta Attention: A Negative Result on Length Extrapolation2026-03-02
FA0194Deterministic Memory Fusion for Long-Horizon Conversational Agents2026-03-02
FA0193Patch, Don't Rewrite: Post-Drift Rule Updates for LogRules-Style LLM Log Parsers2026-03-02
FA0192AR-Order RL Post-Training Reduces Order Robustness in Diffusion Language Models2026-03-02
FA0191HeadRollback: Post-Task Attention Head Rollback for Replay-Free Continual LoRA Fine-Tuning2026-03-02
FA0190Paired Median-of-Means Rewards for Robust Configuration Selection in Vector Search Benchmarking2026-03-02
FA0188Differentially Private Spectral Monitor Logs for Hallucination Detection: A Comparative Study of Wishart and Gaussian Mechanisms2026-03-02
FA0187Differentially Private Eigenspectrum Monitor Logs for Hallucination Detection2026-03-02
FA01868-bit Quantization Provides No Privacy Benefit Against Training-Free Embedding Inversion2026-03-02
FA0184Velocity-Forecast Sampling for Flow-Matching Heads: A Negative Result2026-03-02
FA0181Data-Free Transition-Spectrum Winsorization for Mamba Long-Context Generalization2026-03-02
FA0175Distance-Hiding Fingerprints for Text Embeddings via Secure SimHash2026-03-02
FA0174Action-Support Likelihood Audits Predict Rollout Consistency Failures in Text-Based World Models2026-03-02
FA0172Auditing Norm-Clipped L2-Laplacian Token-Embedding Obfuscation Against Sequence-Aware Reconstruction2026-03-02
FA0171SourceJS-LoRA: Source-Referenced Jensen-Shannon Divergence for Learning LoRA Merge Coefficients2026-03-02
FA0168Token-Balanced Continual Pretraining Eliminates Brain Rot Degradation2026-03-02
FA0163Execution-Signature Recycling: Deduplicating Unit-Test Failure Feedback for Test-Time Code Scaling2026-03-02
FA0162Training-Free Linear Routing for Sparse Attention via Attention-Mass Prediction2026-03-02
FA0161Speaker-Attested Grounding for False Memory Resistance in Agent Memory Systems2026-03-02
FA0156Length-Weighted Loss Does Not Explain the Repetition Advantage in Long-CoT Supervised Fine-Tuning2026-03-02
FA0153Fielded Max-Sim Keying for Assistant-Side Memory Recall in Long-Term Conversational Assistants2026-03-02
FA0151MidPC LoRA: Intermediate SVD Slices for Continual Learning with Low-Rank Adaptation2026-03-02
FA0150ShallowPPL: Investigating Early-Exit Logit Lens for Code Context Compression2026-03-02
FA0149Training-Free Motion-Bias Calibration for Precipitation Nowcasting: A Negative Result2026-03-02
FA0147Quantile Remap Calibration for Precipitation Nowcasting2026-03-02
FA0145Disagreement-Gated Judge KV Reuse: A Training-Free Safety Signal for Multi-Agent LLM Systems2026-03-02
FA0143Tuned-Lens-Style Affine Alignment for Encoder Truncation in Whisper ASR: An Empirical Investigation2026-03-02
FA0142Progress-Guarded LAVE: Lexer-Ignored Stall Filtering for Reliable CFG-Constrained Diffusion Decoding2026-03-02
FA0141BH-Exit: Label-Free Early Termination for HNSW Search via Bucket-Histogram Stability2026-03-02
FA0138Custom Forward-Backward VJPs for DFA-Guided Diffusion Language Models: An Empirical Study2026-03-02
FA0137GaugeFix-LRM: Function-Preserving Q/K Gauge Fixing for Learnable Multipliers in Language Model Training2026-03-02
FA0134Post-hoc Top-$p$ Expert Routing for Dynamic Compute Allocation in Mixture-of-Experts Language Models2026-03-02
FA0131TemplateLeak: A Template-Disjoint Evaluation Audit of CommonForms Form Field Detection2026-03-02
FA0127Budget-Distilled ES-SSM: Cross-Budget Knowledge Distillation for Elastic Spectral State Space Models2026-03-02
FA0123Compute-Matched Repetition Advantage in Long-CoT Supervised Fine-Tuning2026-03-02
FA0122Quote-Batched Payment Protocol for Reducing First-Proposal Bias in Agentic Marketplaces2026-03-02
FA0121Counterfactual Gate Supervision Does Not Fix Gating Credit Assignment in Engram-Style Conditional Memory2026-03-02
FA0116Fact-Check Grounding Loss for Semantically Consistent Model Editing2026-03-02
FA0115OCR-Anchor Reranking: When Best-of-N Selection Fails Due to Candidate Homogeneity2026-03-02
FA0114Sketch-Gated Trace Clustering for Accelerating Inter-Trace Redundancy Pruning2026-03-02
FA0112Interval-Calibrated Noisy Quantization: A Parameter-Free Defense Against Quantization-Gap Attacks2026-03-02
FA0111Label-Free Hyperparameter Calibration for Parallel Context Encoding via KL Divergence Matching2026-03-02
FA0110Targeted Counterfactual Branch Augmentation for Robust Text-Based World Models under Agent Policy Shift2026-03-02
FA0107ConvergeStop: Inference-Time Convergence-Based Halting for Generative Text Embeddings2026-03-02
FA0106TraceBound: Evaluating Trace-Bounded Context for Token-Efficient Coding Agents2026-03-02
FA0105Cross-View PSD Distillation for Viewpoint-Robust Remote Photoplethysmography2026-03-02
FA0104Search-Anchored Hybrid Rollouts for Text-Based World Models2026-03-02
FA0102KL-Time Replay: Function-Space Drift Monitoring for Continual Learning in LLMs2026-03-02
FA0101Task-Aware Early Termination for HNSW via Label-Histogram Stabilization2026-03-02
FA0100Self-Anchored Temporal Filtering for LLM-Free Temporal-Aware Memory Retrieval2026-03-02
FA0087RazorSFT: On-Policy Supervised Fine-Tuning with KL-Minimal Target Selection for Continual Learning2026-03-02
FA0085Tool-Gated Residual Distillation for DataChef Verifier Scoring2026-03-02
FA0083Query-OOD Escalation: Detecting Memory Poisoning Attacks via Embedding-Space Anomaly Detection2026-03-02
FA0082Context Bagging: Inference-Time Ensembling for Robust Long-Context QA Under Hard Distractors2026-03-02
FA0080Misalign@k: Tail-Risk Evaluation of Emergent Misalignment Defenses Under Repeated Sampling2026-03-02
FA0077LogitGate: Probe-Gated Output Logit Bias as a Simplification of Activation Steering for Tool Calling2026-03-02
FA0076Entailment-Checklist Scoring: An API-Free Alternative to LLM-Based Dense Video Caption Evaluation2026-03-02
FA0075Syntax-Diversified Unlearning: Evaluating Data-Side Interventions for Reducing Worst-Case Leakage2026-03-02
FA0074Auditing and Hardening LiveMedBench's Rubric Grader Against Prompt Injection: A Negative Result2026-03-02
FA0073Sink-Free Attention Enables Prefix-Free Streaming KV Caches2026-03-02
FA0072Execution-Trace Guided Remasking for Diffusion Code Generation2026-03-02
FA0069Timeout Bootstrapping for Long-CoT RLVR: Promise and Pitfalls2026-03-02
FA0067Delta-Prefill Switching: Adaptive Routing for Speculative Decoding in Multi-Turn LLM Serving2026-03-02
FA0065Mean-Direction Deflation Reranking for Metric Misuse Repair in Frozen Vector Search2026-03-02
FA0064NLL-Guided Full-Attention Layer Selection for Training-Free Sliding-Window Adaptation2026-03-02
FA0063Clarification Timing Does Not Mitigate Anchoring Bias in Tool-Using LLM Agents2026-03-02
FA0061Entropy Dynamics Do Not Provide Reliable Execution-Free Selection Signals for Code Generation2026-03-02
FA0059Last-Write-Wins Memory: Isolating Deterministic Overwrite Semantics for Long-Context Conflict Resolution2026-03-02
FA0058Chunked Budget Allocation Prevents Non-Monotonic Regressions in World-Model Verification2026-03-02
FA0057LiveMedBench-Ask1: Evaluating Ask-Before-Answer Behavior in Medical LLMs2026-03-02
FA0056Innovation Saturation Does Not Robustify Kalman-Filtered Importance Ratios in LLM Reinforcement Learning2026-03-02
FA0055Decoupling Snapshot Publication from Staleness Tolerance in Distributed GRPO via Lossless Sparse Patches2026-03-02
FA0053Draft De-anchoring Decoding Does Not Mitigate Contextual Drag in LLM Reasoning2026-03-02
FA0052Does MIS-PO Need Ratio-Based Trajectory Selection? A Random-Rejection Mechanism Test2026-03-02
FA0051Toeplitz Block Mixing for Scalable Multi-Head Linear Attention2026-03-02
FA0050R-MEL: Recovering Contrastive Signal from All-Negative Groups via Prefix-Primed Revision2026-03-02
FA0049Premature Speech EOS is Not a Dominant Failure Mode in Qwen2.5-Omni: An Empirical Study of Text-Length-Coupled Audio Stopping2026-03-02
FA0047Canonical Schema Views for Activation Steering Under Tool-Schema Churn: A Negative Result2026-03-02
FA0046QuoteVerify: Inference-Time Quote-Backed Citation Verification for Deep Research Reports2026-03-02
FA0045Hard Examples Beat Easy Examples in Repetition-Heavy Long-CoT Fine-Tuning2026-03-02
FA0044Selective Self-Reference for LLM-as-a-Judge: Using Self-Consistency to Reduce Error Propagation2026-03-02
FA0043Isolated Solve-Then-Judge: A Simple Defense Against Candidate-Response Prompt Injection for Multimodal LLM Judges2026-03-02
FA0042Distilling Bidirectional Embedding Teachers into Streaming-Compatible Causal Students2026-03-02
FA0041MEL-Code: Transferring Meta-Experience Learning to Code RLVR with Unit-Test Rewards2026-03-02
FA0040Typed-DSL Constrained Data Recipes for Higher Executability in DataChef2026-03-02
FA0039Prefix-Ratio GRPO: Improving Gradient Quality for Reinforcement Learning with Verifiable Rewards2026-03-02
FA0038Citation-Consistent Voting for Permutation-Robust Retrieval-Augmented Generation2026-03-02
FA0036EMA-KPO: Simplifying Kalman Policy Optimization with Fixed-Gain Exponential Smoothing2026-03-02
FA0035LASCon: Loop-Aware Scratchpad Condensation for Terminal Agents2026-03-02
FA0034Adaptive Rerank Budgeting for Video-Text Retrieval via Layer-Disagreement Routing2026-03-02
FA0033Interface-Aware Smoke Tests and Deterministic Import Autofix for Feature-Level Coding Agents: A Negative Result2026-03-02
FA0032RC-MemStop: Risk-Controlled Early Stopping for Long-Context Memory Agents2026-03-02
FA0031Evidence-Grounded Constraint Schemas Do Not Improve Medical LLM Guardrails on LiveMedBench2026-03-02
FA0030Answerability-Gain Rewards for Evidence-Label-Free GRU-Mem Gating: An Empirical Investigation2026-03-02
FA0029Output-Space Allocation Costs for Calibration-Guided LLM Compression: An Empirical Study2026-03-02
FA0028Acceptance-Controlled MIS-PO: Adaptive Trajectory Filtering for Stable Off-Policy RLVR Training2026-03-02
FA0027RefSwap: Counterfactual Reference-Swap Verification for Robust LLM Verifiers2026-03-02
FA0025Risk-Controlled Early Exit for Diffusion Language Models2026-03-02
FA0023Answer-Free Self-Referential Critics: Training Solve-Then-Judge VLM Judges with Preference Labels but Without Ground-Truth Answers2026-03-02
FA0022The Repetition Advantage in Long-CoT SFT is a Termination Effect2026-03-02
FA0021Does iGRPO Need a Good Draft? Best-vs-Worst Self-Conditioning Ablation for RLVR Math2026-03-02
FA0020AlignDefTok: Training-Free Transfer of DefensiveTokens via Embedding-Space Alignment2026-03-02
FA0019Step-Down Bridge Guidance Scheduling for Dual-CFG in Video-Audio Diffusion2026-03-02
FA0018Compute-Matched Evaluation of Transform-Augmented GRPO for Mathematical Reasoning2026-03-02
FA0017Copy-Then-Inpaint: Improving Temporal Consistency in Multi-Step GUI Generation via Selective Region Editing2026-03-02
FA0016Query-Conditioned Marginals for OT-Based Context Compression: An Empirical Investigation2026-03-02
FA0015Orthogonal Junk: Gradient-Orthogonality Data Selection for Continual Pre-Training on Low-Quality Data2026-03-02
FA0013Contractive Recurrent Cores for Depth-Extrapolatable Vision-Language-Action Policies: An Empirical Investigation on LIBERO2026-03-02
FA0012Delta-Map Belief Updates for Stable Spatial Revision in Vision-Language Models2026-03-02
FA0011Caption Distillation for ReVision-Style Text-Only MLLM Pretraining: An Empirical Study2026-03-02
FA0008Confidence-Bounded Unit-Test Rewards for Reinforcement Learning from Verifiable Rewards2026-03-02
FA0007WindowScan-Judge: Robust Safety Judging Against Benign-Padding Attacks via Windowed Scanning and Length-Aware Aggregation2026-03-02
FA0006View-Disagreement Escalation for Robust Web-Agent Trajectory Judges2026-03-02
FA0005Selective Delexicalization to Defend Structured-Output LLM APIs from Control-Plane Jailbreaks2026-03-02
FA0004Anytime-CBU: Adaptive Rollout Allocation for Consequence-Based Utility Scoring2026-03-02
FA0003Deflated-RankICIR: Multiple-Testing-Aware Factor Selection for LLM-Driven Alpha Mining2026-03-02
FA0002Adaptive SRE-Mass Cache Sizing for Hybrid Linear Attention2026-03-02
FA0001Canary-Controlled Safe-Data Interleaving for Reducing Emergent Misalignment2026-03-02