INDEX
Every paper Pith has read and judged, in one searchable index.
-
cs.CV arXiv submitted 2026-08-03Hybrid detector-plus-reasoner counts storm damage from drone photos
Huy Quang Ung +5 · “Assessing the Benefits of Combining Advanced Deep Learning Techniques for Post-Disaster Building Damage Assessment from UAV Imagery”
2608.01906 -
cs.CV arXiv submitted 2026-08-03A single photo and a sentence produce a 3D hand-object interaction
Zhenhao Zhang +3 · “PhotoHOI: Synthesizing 3D Hand-Object Interactions from a Single RGB Photograph”
2608.01905 -
cs.AI arXiv submitted 2026-08-03Closed-loop knowledge graph lifts search agents by ~10 points
Zhaoyang Li +8 · “CoEvoKG: Co-Evolving Knowledge Graphs with Self-Evolving Search Agents”
2608.01904 -
cs.HC arXiv submitted 2026-08-03Emotional cues beat species-accurate motion in virtual animals
Kaoru Sumi +1 · “Emotional Expression in Persuasion by Quadruped Virtual Agents: Toward Cross-Species Design Patterns”
2608.01895 -
cs.DC arXiv submitted 2026-08-03Two clock speeds per layer cut LLM serving energy up to 49%
Cunchen Hu +11 · “Energy-Efficient LLM Serving via Disaggregated Attention--FFN and Flexible Frequency Scaling”
2608.01891 -
cs.GT arXiv submitted 2026-08-03Non-atomic values make autobidding equilibria computable fast
Li Chen +2 · “Beyond the PPAD hardness of Auto-bidding Auctions”
2608.01889 -
cs.CV arXiv submitted 2026-08-030.619 dB boost: map chrominance's info gain over intensity
Ya-nan Guan +7 · “Beyond Illumination: A Conditional Mutual Information-Guided Network for Low-Light Image Enhancement”
2608.01886 -
cs.LG arXiv submitted 2026-08-03A cheap pre-filter speeds up anomaly detection up to 4.8x
Zemin Chao +5 · “CARE: A Cascaded Framework for Efficient and Reliable Time Series Anomaly Detection”
2608.01885 -
cs.MM arXiv submitted 2026-08-03Audio agent that calls tools gains 15 points on hard sound tasks
Yuwen Wang +9 · “Hear, Invoke, and Understand: A Skill-Calling Multimodal Agent for Large Audio Language Models”
2608.01881 -
cs.RO arXiv submitted 2026-08-03Prefix conditioning wins for smooth real-time robot control
MotuBrain Team · “World Action Models in Real Time: An Empirical Study of Smooth Execution via Asynchronous Deployment”
2608.01880 -
cs.LG arXiv submitted 2026-08-03Cut few-shot table error 33.8% via Bayesian-network adaptation
Zijian Shen +4 · “LAB-Tab: LLM-Augmented Bayesian Network Adaptation for Few-Shot Tabular Generation”
2608.01879 -
cs.CV arXiv submitted 2026-08-03ViT training spends its new dimensions on class separation
Kaustubh Kapil +1 · “Transformer Geometry Observatory TGO-III: Semantic Geometry Observatory”
2608.01876 -
cs.AI arXiv submitted 2026-08-03One LLM pass now forecasts a time series and explains why
Seunghan Lee +11 · “ReasonCast: Towards Explainable Time Series Forecasting with Reasoning”
2608.01875 -
cs.CY arXiv submitted 2026-08-03Every move in this nuclear-war simulation can be replayed and audited
Glenn Matlin +3 · “No One Wins in Nuclear War: A Social Simulation of Military Decision-making”
2608.01868 -
cs.CL arXiv submitted 2026-08-03Cuts deep-search turns 33% without accuracy loss
Haosi Mo +6 · “CRISP: Critical Step Perception for Training Efficient Deep Search Agents”
2608.01867 -
cs.CL arXiv submitted 2026-08-03One adapter at layer 7 nearly matches full ASR fine-tuning
Darwin Jelestin Muthu +5 · “Analyzing Speech Condition Effects in Dysarthric ASR: A Layer-wise Probing Study”
2608.01865 -
stat.AP arXiv submitted 2026-08-03Ensemble variability sharpens drought forecast bounds
Henri Funk +4 · “Probabilistic Deep Learning for Drought Forecasting: Role of Internal Climate Variability”
2608.01864 -
cs.AI arXiv submitted 2026-08-03Three-stage post-training lifts enterprise QA from 53% to 91.5%
Xiaofeng Shi +8 · “Wnuan: Staged Post-Training for Question Answering over Proprietary Enterprise Knowledge”
2608.01862 -
cs.DC arXiv submitted 2026-08-03Dynamic module assembly beats fixed supernets in federated learning
Jifeng Chen +2 · “FedJigsaw: Multi-Agent Collaborative Model Reassembly for Decentralized Heterogeneous Federated Learning”
2608.01861 -
cs.LG arXiv submitted 2026-08-03Cosine loss fixes the direction errors MSE ignores
Seunghan Lee +11 · “Beyond Magnitude and Shape: A Direction-Aware Loss for Time Series Forecasting”
2608.01857 -
cs.AI arXiv submitted 2026-08-03Masked diffusion lets captions rewrite their own mistakes
Dongwei Sun +5 · “EchoChange: A Diffusion Language Model with Dual Pass Remasking for Factual Remote Sensing Disaster Change Captioning”
2608.01856 -
cs.RO arXiv submitted 2026-08-03Radar-only drone dodges fast thrown balls in dark and smoke
Dnyandeep Mandaokar +1 · “RADAR Perception for Dynamic Obstacle Avoidance onboard small-scale Quadrotor UAVs”
2608.01855 -
cs.IT arXiv submitted 2026-08-03Subarray motion shaves 24 dB off ISAC sensing CRB
Ruiqi Liu +3 · “Movable Subarray-Aided ISAC in Hybrid Near-Far Field Channels”
2608.01854 -
physics.soc-ph arXiv submitted 2026-08-03Preferential attachment is a perception
Alexei Vazquez · “Local network growth: How simple rules drive network complexity”
2608.01853 -
cs.DB arXiv submitted 2026-08-03Hybrid graph-vector search is needed for 3D similarity in digital twins
Oussama Zaid (MALT) +4 · “Multimodal Embeddings for 3D Similarity Search in Semantic Web-of-Things Digital-Twin Platforms”
2608.01852 -
cs.RO arXiv submitted 2026-08-03Three robot systems close the full self-improvement loop
Gaytri Jena +5 · “Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills”
2608.01851 -
cs.AI arXiv submitted 2026-08-03Neural net finds tokamak mode frequency from 6% of field data
Dengdi Sun +7 · “Physics-Informed Neural Networks for Complex Eigenfrequency Identification and Mode Structure Reconstruction of the Ground-State ITG Branch”
2608.01850 -
cs.AI arXiv submitted 2026-08-03SPAR erases unsafe answers without wrecking benign replies
Junxiang You +5 · “Exploring and Bridging Knowledge Holes in Unlearned Multimodal Large Language Models”
2608.01849 -
cs.CV arXiv submitted 2026-08-03OCR scores overstate how much visual compression keeps
Yonghan Gao +5 · “Decoupling semantics from vision: A framework for faithful visual-text compression evaluation”
2608.01848 -
cs.AI arXiv submitted 2026-08-03Relaxing FP4 scale coupling restores 99.8% of FP16 accuracy
Xianglong Yan +6 · “FOCUS: FP4 Optimization via Coupled-Relaxation and Dual-Granularity Scaling”
2608.01847 -
cs.IR arXiv submitted 2026-08-03Layer-by-layer hypergraph repair lifts POI ranking by 8.2%
Jinze Wang +4 · “HyperAgent4POI: Dynamic Semantic Message Passing on Multi-Agent Hypergraphs for Missing-Modality Recommendation”
2608.01846 -
cs.LG arXiv submitted 2026-08-03Latent-step caching hits 4.92x speedup on diffusion world models
Leyang Chen +3 · “WorldDynCache: Risk-Controlled Latent Dynamics Approximation for Diffusion World Model”
2608.01845 -
eess.SP arXiv submitted 2026-08-03Textile bus carries power and 1 Mb/s UART data
Akihito Noda · “UART for Wearables (U4We): DC Power and Carrierless Signal Transfer over Conductive Textiles”
2608.01843 -
cs.DS arXiv submitted 2026-08-03Product-state trace distance: 4.73-approximation
Kun He +3 · “Approximating the Trace Distance Between Product Quantum States”
2608.01841 -
cs.LG arXiv submitted 2026-08-03Operator learning speeds transcranial ultrasound simulation 56,000x
Minjee Seo +4 · “tFUSOperator: Operator Learning for Transcranial Focused Ultrasound Digital Twins”
2608.01839 -
cs.AI arXiv submitted 2026-08-03Persistence-based teacher weights lift ALFWorld success by 15.6
Chunji Lv +8 · “PCSD: Persistent Consistency for Self-Distillation in Agentic Reinforcement Learning”
2608.01837 -
cs.AI arXiv submitted 2026-08-03Fine-tuning rewrites behavior geometry
Juntong Wang +3 · “Rewriting or Reweighting? A Geometric Account in Language Models”
2608.01835 -
cs.RO arXiv submitted 2026-08-03VR-only teleoperation trains humanoid policies to 95% success
Bingqian Wu +4 · “Teleopit: A Full-Embodiment Humanoid Teleoperation System”
2608.01834 -
cond-mat.dis-nn arXiv submitted 2026-08-03Swap Monte Carlo cuts grokking time from 3000 to 650 epochs
Lai Shun Chan +4 · “Tunneling the Loss Landscape: Bypassing Memorization with Monte Carlo Parameter Swapping”
2608.01833 -
cs.CV arXiv submitted 2026-08-03One blind 4K model clears haze
Yongcong Wang +10 · “MoCRA: Mixture of Compositional Rank-1 Atoms for 4K All-in-One Video Restoration”
2608.01829 -
cs.CV arXiv submitted 2026-08-03SFT alone teaches long-horizon vision search
Huanyao Zhang +14 · “DeepVoyager-VL: Incentivizing Vision-in-the-Loop Search for Long-Horizon Multimodal Agents”
2608.01827 -
cs.CV arXiv submitted 2026-08-03Timing visual evidence speeds diffusion VLMs 1.55x
Yongkang Zhou +4 · “DAVET: Denoising-Aware Visual Evidence Trajectory Allocation for Diffusion Vision-Language Models”
2608.01821 -
cs.HC arXiv submitted 2026-08-03PPG glucose models no better than mean guess on new people
Supraja Ramesh +4 · “Reassessing the Feasibility of PPG-Based Non-Invasive Blood Glucose Level Estimation”
2608.01820 -
cs.LG arXiv submitted 2026-08-03Simple LSTM beats complex hybrids on engine-life prediction
Fatih \"Urgen +1 · “Predictive Maintenance: Deep Learning-Based Remaining Useful Life Prediction for Combat Aircraft Engines”
2608.01819 -
cs.CL arXiv submitted 2026-08-03Two senses of one word diverge in predictions
K. Jack Scott +2 · “Divergent large language model predictions from convergent representations in ambiguous word pairs”
2608.01816 -
cs.CL arXiv submitted 2026-08-03Few synthetic probes reveal hidden coupling in LLM judge rubrics
Divyansh Singh +2 · “RADAR: Rubric-Aware Dependency and Redundancy Analysis for LLM-as-Judge Evaluation”
2608.01810 -
cs.CV arXiv submitted 2026-08-03Second-opinion gate checks anatomy net on just 9% of X-rays
Siam Tahsin Bhuiyan +6 · “SecondOpinion: Anatomy-Aware Gated Reasoning for Efficient Medical Image Analysis”
2608.01808 -
cs.CV arXiv submitted 2026-08-03Per-target fusion and calibration top four RGBT benchmarks
Zhaoding Ding +4 · “Parameter-Dynamic Adaptive Fusion and Calibration Network for RGBT Tracking”
2608.01807 -
cs.RO arXiv submitted 2026-08-03Flying robot runs impedance and admittance control at once
Zicheng Luo +5 · “Hybrid Impedance-Admittance Control with Multi-Link Aerial Robot for Contact-Rich Surface Sliding Task”
2608.01800 -
cs.SD arXiv submitted 2026-08-03Four frozen speech models catch every unseen deepfake
Seunghyun Kim +2 · “Multi-Backbone Self-Supervised Ensembles for Audio Deepfake Detection and a Cross-Track Analysis of Generation-Detection Asymmetry”
2608.01796