Pith. sign in

INDEX

Every paper Pith has read and judged, in one searchable index.

92,285 reviewed papers in cs · newest first · page 10

Pith rank means recent reader upvotes with each upvote's ranking weight halved after 24 hours. The archive itself is sorted by arXiv submission date.

  1. eess.SY arXiv submitted 2026-08-03
    Amplify-then-attenuate feedback pairs keep control signals bounded

    Arthut Castello B. de Oliveira +2 · “On input-output persistency and the interconnection of positive nonlinear systems”

    2608.01699
  2. cs.RO arXiv submitted 2026-08-03
    Inertia fix in simulator cuts robot-leg sim-to-real errors by 80%

    Jinsong Hong +4 · “Bridging the Sim-to-Real Gap in Parallel-Link Leg Mechanisms via Simulator-Side Dynamics Normalization”

    2608.01697
  3. cs.CV arXiv submitted 2026-08-03
    Keeping player roles separate lifts action-spotting F1 to 0.778

    Ruifeng Wang +2 · “Entity-Aware Sequence Transduction for Player-Centric Ball Action Spotting”

    2608.01696
  4. cs.HC arXiv submitted 2026-08-03
    No-score VR neck rehab relaxes users

    David Perron +6 · “Beyond Score-Based Gamification: Designing Spatiotemporal and Musical Experiences for VR Neck Rehabilitation”

    2608.01688
  5. cs.CV arXiv submitted 2026-08-03
    Scaling laws point to national medical AI without big tech

    Masahiro Oda · “Generative AI and Foundation Models in Medical Image”

    2608.01686
  6. cs.AI arXiv submitted 2026-08-03
    Below 40%: LLM agents fail most complex graph tasks

    Jiarui Tan +7 · “GABench: A Comprehensive Benchmark for Evaluating LLM Agents on Graph Analysis Tasks”

    2608.01684
  7. cs.AI arXiv submitted 2026-08-03
    Memory consolidation erases authority in 48 of 49 setups

    Qiuyang Zhan +4 · “When Memory Becomes Authority: Benchmarking Authority Collapse at the Memory Consolidation Boundary”

    2608.01679
  8. cs.LG arXiv submitted 2026-08-03
    RL-trained editor writes agent skills by testing each edit

    Junhao Shen +3 · “Progressive Agent Skill Generation via Reinforcement Learning”

    2608.01678
  9. cs.CV arXiv submitted 2026-08-03
    Diffusion bridge turns routine cine MRI into DENSE-quality strain

    Rishov Paul +2 · “Generative Brownian Bridge Diffusion In Motion Space For Enhanced Myocardial Strain Analysis”

    2608.01677
  10. cs.CL arXiv submitted 2026-08-03
    Block pruning rewires which evidence controls a model's answer

    Xingyu Ren +3 · “Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation”

    2608.01676
  11. cs.CL arXiv submitted 2026-08-03
    Distilling what to remember doubles long-context retrieval

    Zixuan Wang +6 · “Learning What to Remember: Test-Time Training via Context Distillation”

    2608.01672
  12. cs.HC arXiv submitted 2026-08-03
    CellPrism narrows thousands of perturbation strategies to one

    Chuhan Shi +5 · “CellPrism: A Visual Analytics System for Exploring AI-Driven Virtual Cells in Drug Discovery”

    2608.01669
  13. cs.NI arXiv submitted 2026-08-03
    LEO AI scheduler trims delay variance at 1% throughput cost

    V\'ictor Vilchez +4 · “LEO-Aware DRL Meta-Scheduler for 5G Non-Terrestrial Network Slicing”

    2608.01668
  14. cs.AI arXiv submitted 2026-08-03
    TCPO turns verifier scores into per-turn credit

    Sicong Liao +2 · “TCPO: Turn-Level Credit Policy Optimization”

    2608.01667
  15. cs.CL arXiv submitted 2026-08-03
    LLM judges let style override idea substance

    Fengxian Ji +10 · “Style Wins, Substance Loses: A Diagnosis of LLM-as-Judge in Idea Generation”

    2608.01666
  16. cs.AI arXiv submitted 2026-08-03
    Split budgets before ranking tokens: audio-video AI keeps 98.7% at 25%

    Zhenghui Guo +7 · “Allocation Before Ranking: Decoupled Token Compression for OmniLLMs”

    2608.01665
  17. cs.CV arXiv submitted 2026-08-03
    Causal voxel cache keeps splat quality climbing to 1024 views

    Changhao Song +4 · “StreamSplat: Streaming Feed-Forward 3D Gaussian Splatting”

    2608.01659
  18. cs.DC arXiv submitted 2026-08-03
    Cache shield lifts victim hit rate from 5% to 85%

    Zhiyu Wang +1 · “Preserving Admission Responsibility in Multi-Tenant Large Language Model Prefix Caches”

    2608.01657
  19. cs.HC arXiv submitted 2026-08-03
    Trajectories, not snapshots, for evaluating adaptive AR

    Nuwan Janaka +4 · “You Cannot Optimize What You Cannot Measure: Multitasking Evaluation as the Missing Foundation of AI-Mediated Heads-Up Interaction”

    2608.01656
  20. cs.DC arXiv submitted 2026-08-03
    Prefix cache placement hits ~99.8% of optimal

    Zhiyu Wang +1 · “PrefixPlace: Provable Prefix Key-Value Placement for Large Language Model Serving under Heterogeneous Compute and Transfer Costs”

    2608.01655
  21. cs.RO arXiv submitted 2026-08-03
    One-shot plans beat re-planning LLMs at 0.05% of runtime

    Shen You +19 · “SyncPlan: Long-Horizon LLM Coordination with Explicit Synchronization and Adaptive Correction”

    2608.01652
  22. cs.DC arXiv submitted 2026-08-03
    A tree of draft tokens now verifies in one parallel GPU pass

    Li Wang +11 · “Bole: Efficient Tree Speculation for Hybrid-Attention Language Models”

    2608.01651
  23. cs.NI arXiv submitted 2026-08-03
    LLM-generated routing rewards hit within 3% of expert baseline

    Walter P. Casas +2 · “LLM-Driven Automated Reward Design for Reinforcement Learning-Based Routing in LEO Satellite Networks”

    2608.01649
  24. cs.LG arXiv submitted 2026-08-03
    Hardware-error forecasting works only for stable error patterns

    Kaiyuan Liao +5 · “Evaluating Forecasting Techniques for Hardware Errors on a Large-scale HPC System”

    2608.01648
  25. cs.NE arXiv submitted 2026-08-03
    Spiking transformer reads handwriting in just two timesteps

    Xiubo Liang +5 · “Spike-HTR: Spiking Neural Transformer for Handwritten Text Recognition”

    2608.01646
  26. cs.AI arXiv submitted 2026-08-03
    Best LLM agent solves 32.7% of real GIS tasks

    Abhinav Pothuri +3 · “GISAgentBench: A Practitioner-Sourced Benchmark for Evaluating LLM Agents on GIS Tasks”

    2608.01645
  27. cs.CV arXiv submitted 2026-08-03
    CRAFT keeps 97% of video-VLM accuracy at 8x compression

    Yu Chen +5 · “CRAFT: Compression via Recursive Adaptive Fusion of Video Tokens for Vision-Language Models”

    2608.01644
  28. cs.CR arXiv submitted 2026-08-03
    Machine-learning malware detector hits 98.6 percent accuracy

    ABM.Adnan Azmee +4 · “Performance analysis of Machine learning algorithms for predicting malware”

    2608.01642
  29. cs.LG arXiv submitted 2026-08-03
    GraphIR state lifts AI-guided architecture search

    Zhen Liu +5 · “GraphIR: Architecture-Level Search States for LLM-Guided Neural Architecture Evolution”

    2608.01633
  30. cs.CL arXiv submitted 2026-08-03
    AI cache compression can keep answers right but wreck reasoning

    Mengting Ai +2 · “Does Accuracy Equal Evidence? Reasoning Faithfulness under KV Cache Compression”

    2608.01631
  31. cs.CL arXiv submitted 2026-08-03
    Learned search inside the weights matches RAG at 3–19x lower latency

    Shicheng Xu +9 · “RING: Retrieval-Internalized Generation for Continual Large-Scale Knowledge Injection”

    2608.01630
  32. cs.CL arXiv submitted 2026-08-03
    Six LLMs reproduce Japan's bias against non-native writing

    Naho Orita +2 · “Human-LLM Alignment in Language Attitudes Toward Non-Native Japanese”

    2608.01629
  33. cs.CV arXiv submitted 2026-08-03
    No pose needed: motion jumps from kangaroos to lamps

    Zhixue Fang +8 · “Motion Beyond Morphology: Bootstrapping Cross-Category Motion Transfer from Abstract Motion Representations”

    2608.01628
  34. cs.LG arXiv submitted 2026-08-03
    Scale-split bottleneck keeps rain forecasts sharp for 60 minutes

    Zhuo Wang +3 · “QWRF-Net: A Quantum-Wavelet Framework with Rectified Flow for Short-Term Precipitation Nowcasting”

    2608.01626
  35. cs.CL arXiv submitted 2026-08-03
    One knob governs gradient-free LLM adaptation

    Taeyeong Kim +3 · “Not the Dimension, the Norm: What Matters in Gradient-Free Weight Perturbation of Language Models”

    2608.01624
  36. eess.SP arXiv submitted 2026-08-03
    8.67 dB SI-SDR: soft EEG gate tracks attention switches mid-trial

    Xuefei Wang +4 · “SAGE: Switch-Aware EEG-Guided Soft Gating for Target Speaker Extraction with In-Trial Switching”

    2608.01623
  37. cs.NE arXiv submitted 2026-08-03
    Spiking Transformer matches ANN accuracy

    Xiubo Liang +5 · “SMM Transformer: Leveraging Spiking Neural Networks for Multimodal Tasks”

    2608.01622
  38. eess.SP arXiv submitted 2026-08-03
    Adaptive gate lifts EEG switch detection to 67% F1

    Yuting Ding +4 · “SGAD: A State-Guided Adaptive Decision Framework for Robust EEG-Based Auditory Attention Switch Decoding”

    2608.01618
  39. cs.GT arXiv submitted 2026-08-03
    Four unit-conjugate primes give four different shuffles

    Matthew Fried · “Conjugation invariants determine the metacommutation permutation only up to relabelling”

    2608.01610
  40. cs.CR arXiv submitted 2026-08-03
    Rug-pull detector reaches 92.7% accuracy on meme coins

    Jinyin Song +2 · “From Viral to Void: Multi-Dimensional Behavioral and Contractual Analysis for Rug Pull Identification”

    2608.01609
  41. math.CO arXiv submitted 2026-08-03
    300 ones in a 43x43 grid force an all-one 2x2 block

    Ankan Sadhu · “An Improved Upper Bound on the Zarankiewicz Number z(43;2)”

    2608.01606
  42. cs.GR arXiv submitted 2026-08-03
    Pretrained speech encoder beats top 3D talking-head models

    Jiu-Cheng Xie +6 · “ETHead: Generating Expressive 3D Facial Animation and Head Movement from Speech”

    2608.01605
  43. cs.RO arXiv submitted 2026-08-03
    Moving the contact point with the object lifts robot success to 70%

    Gaoyuan Wu +4 · “AffordTrajDP: Dynamic Affordance-Guided Visuomotor Policy Learning for Robotic Manipulation”

    2608.01603
  44. cs.CV arXiv submitted 2026-08-03
    EF 'calibration gain' traced to a 6.3-point measurement gap

    Dang P. M. Cao +1 · “When Measurement Conventions Masquerade as Calibration Gains in Cardiac Digital Twins”

    2608.01602
  45. cs.RO arXiv submitted 2026-08-03
    Humanoid robot replays eight construction actions from video demos

    Yanxi Liu +1 · “Perception-and-action system for humanoid robot task execution in construction”

    2608.01600
  46. cs.LG arXiv submitted 2026-08-03
    Average accuracy hides where volatility forecasts fail

    Arthur Chagas +5 · “Latent-Regime Bias Auditing for Volatility Forecasting”

    2608.01599
  47. cs.CL arXiv submitted 2026-08-03
    Revealing characters' ignorance lifts LLM false-belief accuracy 7.3%

    Eojin Jeon +1 · “PICTURE: Enhancing Theory-of-Mind in Large Language Models by Revealing, Not Hiding, Characters' Lack of Knowledge”

    2608.01598
  48. cs.AI arXiv submitted 2026-08-03
    Structured problem maps beat full traces in self-distilled reasoning

    Xuyang Zhao +6 · “Is More Privileged Information Better? From Solution Traces to Problem-Solving Structure in Self-Distilled Reasoning”

    2608.01589
  49. cs.CV arXiv submitted 2026-08-03
    Dual-depth priors lift sparse-camera Gaussian rendering by 1.33 dB

    Jijian Zhao · “D^2-4DGS: Dual-Depth Guided Sparse-Camera 4D Gaussian Splatting”

    2608.01588
  50. stat.ML arXiv submitted 2026-08-03
    Temporal-aggregate labels cap snapshot R² at a trait-determined floor

    Xizhe Zhang · “The Label Defines the Timescale: Trait-State Limits of Temporal-Aggregate Learning”

    2608.01587