INDEX
Every paper Pith has read and judged, in one searchable index.
-
q-bio.NC arXiv submitted 2026-07-14Fall detection as continuous stability loss on a dual-LTC edge network
Wenjun Xia +3 · “Real-time fall detection based on vision for low-power edge platforms”
2607.12909 -
cs.CV arXiv submitted 2026-07-14Rank-1 consensus beats score thresholds for face enrollment
Gabriella Pangelinan +3 · “Rank-1 Identity Consensus Predicts Gallery Enrollment in 1:N Face Matching More Accurately than Score Thresholding”
2607.12903 -
stat.ML arXiv submitted 2026-07-14Randomized HMC mixes in O(√(1/α) log 1/ε) total time
Siddharth Mitra +3 · “Accelerated Mixing Time of Randomized Hamiltonian Monte Carlo”
2607.12902 -
cs.AR arXiv submitted 2026-07-1432-channel bio-chip adapts event rate to neural-signal envelope
Narayanan Shyam +2 · “A 32-channel event-based bio-signal analog front-end with adaptive delta and pulse frequency encoding”
2607.12901 -
cs.CV arXiv submitted 2026-07-14One Transformer maps visual examples
Yunzhou Li +7 · “UniMedSeg: Unified In-Context Learning for Multi-Paradigm 2D/3D Medical Image Segmentation”
2607.12896 -
cs.CV arXiv submitted 2026-07-143B-activated model nears 32B-class embodied agent scores
Ziyi Wang +21 · “Hy-Embodied-VLM-1.0: Efficient Physical-World Agents”
2607.12894 -
cs.AI arXiv submitted 2026-07-14Memory fails in five ways
Xixuan Hao +8 · “MemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversations”
2607.12893 -
cs.RO arXiv submitted 2026-07-14Retrieval across demos fixes time-based robot progress labels
Lirui Zhao +5 · “UR-VC: Unsupervised Robotic Value Correction for Time-Derived Progress Proxies”
2607.12892 -
cs.LG arXiv submitted 2026-07-14Energy losses teach nets to find tensegrity shapes and forces
Jing Qin +1 · “Energy-Based Physics-Informed Form Finding for Clustered Tensegrity Structures”
2607.12888 -
cs.AI arXiv submitted 2026-07-14Agents write symptom extractors that beat lexicons without fine-tuning
Cameron Cagan +5 · “A Multi-Agent System for Autonomous, Fine-Tuning-Free Clinical Symptom Detection: Development and Validation Study”
2607.12886 -
cs.CL arXiv submitted 2026-07-14LLM judges flip up to 85% of verdicts once a reference appears
Chalamalasetti Kranti +1 · “LLM Judges Can Be Too Generous When There Is No Reference Answer”
2607.12885 -
cs.CL arXiv submitted 2026-07-14Medical AI fixes myths at 85%
Monica Munnangi +1 · “Evaluating Large Language Models on Misconceptions in Multi-Turn Medical Conversations”
2607.12884 -
cs.MM arXiv submitted 2026-07-14Personalized thumbnails beat generic ones in clicks
Zhiyu He +5 · “What Would You Click? Personalized Video Thumbnail Generation with Preference-aware Highlight Retrieval”
2607.12882 -
cs.CV arXiv submitted 2026-07-14Negative attention scores help ViTs ignore backgrounds
Peter R.D. van der Wal +2 · “Inhibited Self-Attention: Sharpening Focus in Vision Transformers”
2607.12881 -
cs.MA arXiv submitted 2026-07-14LLMs compile custom inference engines from constraints alone
Zhenwen Miao +2 · “MetaInfer: A Knowledge Only LLM Inference Engine Generator SKILL Toolbox”
2607.12875 -
cs.CV arXiv submitted 2026-07-14Metrics that score synthetic images predict better zero-shot detectors
Martina Radoynova +4 · “Metric-Guided Synthetic Image Data Rendering for Deep Learning compatible with Agentic AI”
2607.12874 -
cs.SD arXiv submitted 2026-07-14Real-time music enhancement works—but can also make things worse
Emmanouil Karystinaios +4 · “Low-Latency Neural Models for Real-Time Music Enhancement”
2607.12872 -
cs.IT arXiv submitted 2026-07-14Survey maps code equivalence attacks and open hardness gaps
Anna-Lena Horlemann +3 · “A Survey on Code Equivalence: The State-of-the-Art and Open Questions”
2607.12870 -
cs.SE arXiv submitted 2026-07-1414,227 labeled DNN runs turn training curves into fault detectors
Sigma Jahan · “Deep4ge: DNN Training Trajectories for Fault Detection and Diagnosis”
2607.12868 -
cs.GT arXiv submitted 2026-07-14Closed-form equilibria for quiz-show prize-budget games
Mingshi Yao +2 · “Quiz Show Games: Searching with Bimodal Hiding”
2607.12867 -
cs.CV arXiv submitted 2026-07-14Sigmoid loss stops models from reconstructing anomalies
Nguyen Minh Tri +2 · “Statistical Non-linear Reconstruction Loss for Image Anomaly Detection”
2607.12866 -
cs.SE arXiv submitted 2026-07-14ROBIN shrinks WinoBias gaps by editing head subspaces
Sigma Jahan · “Toward Localizing and Repairing Bias in Transformer Attention Heads”
2607.12863 -
cs.ET arXiv submitted 2026-07-14Two Ising maps of ML decoding create opposite neuromorphic costs
George N. Katsaros +2 · “A Comparative Analysis of Ising Formulations for Neuromorphic Maximum-Likelihood Channel Decoding”
2607.12862 -
cs.RO arXiv submitted 2026-07-14Simple rewards make robot swarms chase hidden geometry
Yize Mi +3 · “Unveiling Complex Collective Behaviors from Simple Rewards”
2607.12861 -
cs.CV arXiv submitted 2026-07-14Road detectors learn layout rules
Shiyi Mu +2 · “LARAD: Layout-Aware Road Anomaly Detection via Spatial-Logic Reasoning”
2607.12858 -
cs.SD arXiv submitted 2026-07-14Seven axes pass corruption tests for free-note chart feedback
Jhen-Ke Lin · “ChartGenEval: Corruption-Tested Multi-Dimensional Feedback for Rhythm-Game Chart Generation”
2607.12857 -
cs.LG arXiv submitted 2026-07-14DP rewards fine-tune a reasoning model to near-optimal TES control
Takumi Shioda +2 · “Verifier-Based Reinforcement Fine-Tuning of Reasoning Models for Thermal Energy Storage Control”
2607.12856 -
cs.HC arXiv submitted 2026-07-14No-code system turns every click into a valid graph query
Michael Behrisch +3 · “GraphPolaris: A System for Query, Analysis, and Visualization of Graph Databases”
2607.12845 -
cs.ET arXiv submitted 2026-07-14Mixed-size nanodots keep magnetic reservoirs stable from 5–35 °C
Zhengfei Chen +5 · “Reproducible Reservoir Computing with Thermally Driven Superparamagnets: Controlling Temperature Sensitivity”
2607.12840 -
cs.DC arXiv submitted 2026-07-14Edge LLMs run 1.7× faster by co-scheduling iGPU and NPU
Gregory Hyegang Jun +6 · “HeteroMosaic: Exposing and Exploiting Heterogeneous Execution Opportunities for Energy-Efficient Edge LLM Inference”
2607.12839 -
cs.CL arXiv submitted 2026-07-14LLM rubrics nearly match human scores for paper reproduction
Hanhua Hong +6 · “Can LLMs Write Reliable Rubrics? A Meta-Evaluation for Experiment Reproduction”
2607.12835 -
stat.ML arXiv submitted 2026-07-14One neural map learns full circular conditionals for pose and wind
Rajdeep Pathak +2 · “ANGLE: Angular Neural Generative Learning via Engression”
2607.12833 -
cs.CL arXiv submitted 2026-07-14Hide named entities at train time
Roi Cohen +7 · “Knowledgeless Language Models: Suppressing Parametric Recall for Evidence-Grounded Language Modeling”
2607.12831 -
cs.LG arXiv submitted 2026-07-14Why parallel diffusion LLMs still need special tricks to win on latency
Daehoon Gwak +3 · “Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques”
2607.12829 -
quant-ph arXiv submitted 2026-07-14Quantum ensemble cuts phishing misses on Ethereum
Sai Sakunthala Guddanti +3 · “Detecting Phishing in Ethereum Networks using Quantum Machine Learning”
2607.12828 -
cs.AI arXiv submitted 2026-07-14Causal rules that stay maximally specific end Hempel's ambiguity
Evgenii Vityaev · “Solution of the Hempel's statistical ambiguity problem and Causal AI”
2607.12826 -
cs.AI arXiv submitted 2026-07-14AI chats train your brain: same session
Eranga Bandara +19 · “Human-AI Agent Interaction as a Neuroplastic Training Environment”
2607.12823 -
cs.CV arXiv submitted 2026-07-147B model learns to bind sound effects to visual events
Yanghai Wang +11 · “AVSCap: Orchestrating Audio-Visual Synergy for Omni-modal Video Captioning”
2607.12820 -
cs.CV arXiv submitted 2026-07-14VLMs audit place matches, lifting recall 13.6% without fixed thresholds
Sania Waheed +3 · “Breaking D\'ej\`a Vu: Independent Auditing of Visual Place Recognition through Vision-Language Reasoning”
2607.12818 -
cs.AI arXiv submitted 2026-07-14CoT mostly reuses early visual features
Hiroto Osaka +5 · “Visual Access Boundaries in Vision-Language Model Reasoning”
2607.12815 -
cs.RO arXiv submitted 2026-07-14Pixel loops cut navigation failures by 35 points
Sarthak Chittawar +6 · “PixelLoop: Shortcut Topological Navigation with Pixel-Level Loops”
2607.12811 -
math.NA arXiv submitted 2026-07-14Least-squares weak Galerkin yields optimal Fokker-Planck errors
Chunmei Wang +1 · “A Least Squares Weak Galerkin Finite Element Method for Fokker-Planck Type Equations”
2607.12803 -
cs.RO arXiv submitted 2026-07-14Fixed-wing drones track and hit moving targets without self-occlusion
Wei-Hao Liou +1 · “Autonomous Tracking and Terminal Guidance of Moving Targets for Fixed-Wing UAVs”
2607.12801 -
cs.CV arXiv submitted 2026-07-14AI reasons in pure pixels, jumps 25% on visual tasks
Zhongwei Ren +7 · “UniVR: Thinking in Visual Space for Unified Visual Reasoning”
2607.12800 -
cs.CL arXiv submitted 2026-07-1444 models chose “serendipity” 41% of the time
Tapan Parikh · “The One-Word Census: Answer-Choice Conformity Across 44 Language Models”
2607.12796 -
cs.CR arXiv submitted 2026-07-14Judge-free SafetyAUC ranks internal danger recognition
Roman Prosvirnin +3 · “Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels”
2607.12792 -
cs.AI arXiv submitted 2026-07-14Co-evolved metric recovers 88–110% of skill gains without a true grader
Xing Zhang +6 · “Who Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents”
2607.12790 -
cs.AR arXiv submitted 2026-07-143D floorplanner optimizes for real BIPS
Shuo Ren +8 · “CLIP-3D: Closed-Loop Evaluation of Performance and Physical Constraints for 3D ICs”
2607.12788 -
cs.LG arXiv submitted 2026-07-14Attention codebooks refine only where the mass is
Winfried van den Dool +4 · “AVQ-Attention: Adaptive Vector-Quantized Attention”
2607.12789 -
cs.AI arXiv submitted 2026-07-14Sub-1B model matches large MER systems on nine benchmarks
Kaiwen Zheng +5 · “Do We Really Need Multimodal Emotion Language Models Larger Than 1B Parameters?”
2607.12787