INDEX
Every paper Pith has read and judged, in one searchable index.
-
cs.LG arXiv submitted 2026-07-31Physics law inside the network cuts battery health forecast error
Zeping Chen +5 · “PiDDM: Physics-Informed Differentiable Degradation Modeling for Lithium-Ion Battery State-of-Health Prediction”
2607.29095 -
q-bio.NC arXiv submitted 2026-07-31One Python function computes all 17 metacognition measures
Saurabh Ranjan +3 · “metasignal: A Python Package for Comprehensive Metacognitive Analysis and Decision-Making”
2607.29093 -
cs.LG arXiv submitted 2026-07-31190 studies: surgical ML runs on private data
Yizhi Dong +6 · “What Is Missing in Surgical Risk Stratification and Outcome Prediction: A Scoping Review of End-to-End Machine Learning Approaches”
2607.29090 -
cs.CY arXiv submitted 2026-07-31Deployment friction, not model IQ, decides AI success
Fabricio F. Costa (HCLTech) · “The Deployment Wall: A Diagnostic Framework and Instrument for Enterprise AI in the Deployment Era”
2607.29089 -
cs.AI arXiv submitted 2026-07-31Three open-source AI models in relay match GPT-5.2 at 1/7 the cost
Yanbin Fang +2 · “Harnessing the Wisdom of LLM Crowds through Complementarity-Driven Iterative Collaboration”
2607.29087 -
cs.SD arXiv submitted 2026-07-31Music AI models embed pitch as a helix
Hayato Yagi +4 · “Do Music Foundation Models Embed Pitch in Helical Structure?”
2607.29086 -
cs.CY arXiv submitted 2026-07-3177 of 100 emergencies get downgraded despite a 100% safety score
Anthonio Oladimeji Gabriel +1 · “IyawoBench v2.0: Extended Diagnostic Evaluation of Large Language Model Clinical Triage in Nigerian Primary Care”
2607.29085 -
cs.CV arXiv submitted 2026-07-31184-D rig controls recognize walkers at 67.8% Rank-1
Huiran Duan +6 · “MHRGait: Gait Recognition from Momentum Human Rig Pose”
2607.29083 -
cs.CL arXiv submitted 2026-07-31Zero-shot LLM matches trained model at spotting child stunting
Muhammad Ashad Kabir +1 · “Can Zero-Shot LLMs Predict Child Malnutrition? A Fairness and Temporal Robustness Study”
2607.29082 -
cs.CL arXiv submitted 2026-07-31Refresh interval, not confidence threshold, controls AI content drift
Yaoxuan Dou +1 · “Faster but Different: Diagnosing and Controlling Content Drift in Accelerated Multimodal Diffusion Language Models”
2607.29079 -
cs.LG arXiv submitted 2026-07-31Teacher help arrives only after the student drifts
Yuchen Xia +5 · “DASH-OPD: Discrepancy-Aware Switching with Hysteresis for On-Policy Distillation”
2607.29078 -
cs.AI arXiv submitted 2026-07-31Proof: distance-minimizing counterfactuals are Bayesian MAP estimates
Keita Kinjo · “A Generalized-Bayes Perspective on Counterfactual Explanations: Posterior-Based Decision-Making and Evaluation”
2607.29077 -
cs.AR arXiv submitted 2026-07-31Protecting 5% of key-value memory nearly restores analog LLM quality
Yuannuo Feng +7 · “Selective KV Cache Protection for Noise-Resilient LLM Inference on Analog Compute-In-Memory Systems”
2607.29076 -
eess.SY arXiv submitted 2026-07-31ML classifier cuts average loss 550 kW in solar plant
Qianxue Xia +2 · “Machine Learning-Assisted Stability Boundary Determination of Multiport Autonomous Reconfigurable Solar Power Plants”
2607.29072 -
cs.LG arXiv submitted 2026-07-31Half-memory clients can jointly fine-tune a full-scale model
Shengkun Zhu +7 · “Federated Foundation Models Fine-Tuning with Heterogeneous Compressed Clients”
2607.29071 -
cs.DB arXiv submitted 2026-07-31Query-density partitioning balances PIM load and speeds range queries
Takato Hideshima +2 · “Query Density-Driven Partitioning for Spatiotemporal Load Balancing on Processing-in-Memory Systems”
2607.29070 -
cs.DC arXiv submitted 2026-07-31Tool execution drives up to 48% of agent task latency
Leonid Kondrashov +16 · “Rethinking AI Cloud Infrastructure for Agentic Serving Systems with the Aries Experimentation Framework”
2607.29069 -
cs.CL arXiv submitted 2026-07-31LLMs win on scarce legal data
Theekshana Samaradiwakara +2 · “Semantics of Subterfuge: Benchmarking Legal Deception Detection Against General-domain State-of-the-Art”
2607.29066 -
cs.CL arXiv submitted 2026-07-31Byte-level hashing frees Engram memory from tokenizer lock-in
Jia Peng Lim +1 · “Tokenizer-Agnostic Engram Module”
2607.29065 -
cs.LG arXiv submitted 2026-07-31Frontier LLMs don't beat a keyword script at crash coding
Sudhir Bharati +2 · “Benchmarking Frontier Large Language Models Against Official Crash Database Coding Using Police Crash Narratives”
2607.29064 -
math.NA arXiv submitted 2026-07-31Owen scrambling ruins QMC point spacing
Kosuke Suzuki · “Separation properties of scrambled digital nets and related random point sets”
2607.29063 -
cs.AI arXiv submitted 2026-07-31Evaluation setting, not training data, decides steering effect
Matthew Nguyen +3 · “On the Generalization of Steering Vectors for Chain-of-Thought Faithfulness”
2607.29062 -
cs.CV arXiv submitted 2026-07-31Boost refined-mask IoU by up to 21 points with adversarial noise
Beomyoung Kim +1 · “Learning from Adversity: Semantic-Aware Mask Refinement through Adversarial Perturbation”
2607.29059 -
cs.AI arXiv submitted 2026-07-31Zooming beats whole pages by 10.6 points in one test
Rashid Mushkani +2 · “Evidence-Grounded Constraint Checking in Construction Documents”
2607.29058 -
cs.LG arXiv submitted 2026-07-31Search-based repair boosts multi-agent success by up to 12.1%
Hanxiao Lu +1 · “Autonomous Repair for Multi-Agent Systems via Monte-Carlo Tree Search”
2607.29055 -
cs.LG arXiv submitted 2026-07-31Conformal rules certify local model winners at a controlled error rate
Yi Zhou +3 · “Who Wins Where? Conformal Model Comparison for Local Superiority”
2607.29053 -
cs.RO arXiv submitted 2026-07-31Text reasoning boosts autonomous driving accuracy ~24%
Zeyu Dong +3 · “Outcome-Guided Distillation: A Teacher-Student Framework to Advance VLM Reasoning in Autonomous Driving”
2607.29052 -
cs.LG arXiv submitted 2026-07-31Lookahead lemmas prove 34% more nets safe
Liam Davis +1 · “Learning Lookahead Lemmas for Neural Network Verification”
2607.29051 -
cs.CV arXiv submitted 2026-07-31Spiking 3D models beat full retuning with 5% of parameters
Zihao Guo +4 · “Parameter-Efficient Fine-Tuning for Spiking Point Cloud Models”
2607.29048 -
cs.CV arXiv submitted 2026-07-31Soft emotion graph lifts EmVidCap caption scores by up to 9 points
Junbo Wang +4 · “Adaptive Emotional Video Captioning via Affective Heterogeneous Graph Reasoning and Multi-task Joint Learning”
2607.29045 -
cs.CL arXiv submitted 2026-07-3197% F1: AI compiles ancient commentaries across five editions
Ke Liang +2 · “From Inline Notes to Collected Commentaries: Toward Context-Preserving Organization of Exegetical Knowledge in Classical Chinese Texts”
2607.29044 -
q-bio.GN arXiv submitted 2026-07-31Sparse 'bad' reference boosts synthetic scRNA-seq fidelity
Yu Song +3 · “Improving scDiffusion with Sparsity-Biased Classifier-Free Guidance”
2607.29043 -
cs.IT arXiv submitted 2026-07-31Sum or product of a random variable exceeds 8/7 of its entropy
Rupert Li · “The Entropic Sum-Product Phenomenon”
2607.29042 -
cs.CV arXiv submitted 2026-07-31Four calibrated confidences replace one box score
Juyong Lee +4 · “Rethinking Detection Calibration: A Coordinate and Direction Perspective”
2607.29040 -
cs.CV arXiv submitted 2026-07-31Report semantics lift OCT anomaly detection by 8.8 AUROC
Zihan Nie +6 · “ReMoE: Report-Guided Mixture-of-Experts for Multimodal OCT/OCTA Anomaly Detection”
2607.29039 -
cs.LG arXiv submitted 2026-07-31Mittag-Leffler layer certifies its own series error
Ning Hu +3 · “DFSC: Error-Controlled Differentiable Mittag-Leffler Propagation for Fractional Scientific Machine Learning”
2607.29038 -
cs.CV arXiv submitted 2026-07-31Render-entropy view selection lifts reconstruction quality
Yan Song +5 · “GO-PRE: Goal-Oriented Next-Best-View Selection via Predictive Rendering Entropy for Active 3D Reconstruction”
2607.29037 -
cs.LG arXiv submitted 2026-07-31Koopman upsampling halves Lorenz-63 coefficient error
Pongpisit Thanasutives +1 · “Dynamics-aware identification of governing equations from sparse and noisy data”
2607.29036 -
eess.SY arXiv submitted 2026-07-31Tune one parameter and Oja's flow tracks rotating subspaces
Kentaro Ohki · “On the Oja-Flow-Based Low-Rank Approximation of Kalman-Bucy Filters for Linear Time-Varying Systems”
2607.29034 -
cs.CV arXiv submitted 2026-07-31Two tiny modules lift SAM to 3D and SAM2 to 4D
Yu Song +4 · “SAM+D: Parameter-Efficient Dimensional Lifting of SAM-Family Models via Depth-Routed LoRA and Depth Shifting”
2607.29033 -
cs.MA arXiv submitted 2026-07-31Sparse hidden states improve long-context QA by up to 29 F1
Haodong Lei +6 · “TransMem: Transforming Hidden States into Memory for Large Language Models”
2607.29032 -
cs.RO arXiv submitted 2026-07-31Plan driving by predicting intent
Jiwei Yang +3 · “Auto-JEPA: A Latent World Model of Continuous Intent for End-to-End Autonomous Driving”
2607.29031 -
cs.CV arXiv submitted 2026-07-31Verified rewards lift multi-reference edit win rate to 67%
Yingmao Miao +7 · “Evaluation-Verification Reward for Consistent Multi-Reference Image Editing”
2607.29025 -
math.NA arXiv submitted 2026-07-31Loop products pin matrix families up to basis change
Dexuan Zhou +3 · “Stable Recovery of Matrix Gauge Classes from Pointwise Invariants”
2607.29021 -
math.PR arXiv submitted 2026-07-31Lopsided coin flips resist entropy-reducing symmetrizers
Mokshay Madiman +1 · “Entropic Symmetrization Resistance”
2607.29020 -
cs.CR arXiv submitted 2026-07-31Threshold selection replaces encrypted ranking in RAG
Yang Gao +5 · “GoldenRetriever: Non-Interactive Homomorphic Encrypted Retrieval for Privacy-Preserving RAG”
2607.29019 -
eess.SY arXiv submitted 2026-07-31eVTOL flight noise, mapped as resident annoyance, peaks during descent
Swapnil Saha +2 · “An ArcGIS Framework for Mapping Human-Centered Noise Annoyance for AAM Infrastructure Planning”
2607.29017 -
cs.DS arXiv submitted 2026-07-31New dynamics sample fixed-density independent sets in O~(n/eps)
Mehrad Abbaszadeh Minab +4 · “Nonlinear Exchange Dynamics for Independent Sets”
2607.29016 -
cs.RO arXiv submitted 2026-07-31Airborne jump corrections cap at ~13°/s
Yu Hu +9 · “DART: Dual-Axis Airborne Reachability-Gated Torque-Reaction for Off-Road Vehicle Jumps”
2607.29011 -
cs.IR arXiv submitted 2026-07-3127% better recommendations from self-evolving reasoning distillation
Zhuang Zhuang +6 · “EvoReason: Self-Evolving Reasoning Primitive-Guided On-Policy Distillation for Latent Reasoning in Generative Recommendation”
2607.29010