Pith. sign in

INDEX

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

100,929 reviewed papers in cs · newest first · page 14

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. cs.AI arXiv submitted 2026-08-03
    Five training instructions lift small-model math reasoning 5%

    Jim Dilkes +2 · “Instruction-Conditioned Exploration for Reinforcement Learning with Self-Distillation to an Unconditioned Policy”

    2608.02087
  2. cs.SE arXiv submitted 2026-08-03
    Call-graph context boosts code matching

    Samuel Valenzuela +1 · “Pretraining on Call Graphs: When Binary Analysis Tasks Profit From Context”

    2608.02084
  3. cs.LG arXiv submitted 2026-08-03
    Four-state EEG decoder triggers exoskeleton steps in 70 ms

    Shantanu Sarkar +2 · “A 2-Block Architecture for Real-Time EEG Gait Decoding: A Pilot Study”

    2608.02083
  4. cs.NI arXiv submitted 2026-08-03
    O-RAN model: edge cuts latency

    Urooj Tariq +5 · “Energy-Latency Trade-offs in O-RAN with Distributed Baseband Processing and AI Inference”

    2608.02082
  5. cs.LG arXiv submitted 2026-08-03
    Learn the link, not just the ratings: isotonic Bradley-Terry wins

    Ryoya Yamasaki · “Isotonic Bradley-Terry Model for Paired Comparison Data”

    2608.02081
  6. cs.RO arXiv submitted 2026-08-03
    Printed EIT skin finds touch on curved surfaces within 6 mm

    Haofeng Chen +8 · “Toward Geometry-Scalable Whole-Body Touch for Humanoids: A 3D-Printed Conformal EIT Skin”

    2608.02080
  7. cs.RO arXiv submitted 2026-08-03
    Safety filter keeps drone vision from losing 3D features mid-flight

    Ege C. Altunkaya +3 · “TANGO-VIO: Triangulation-Aware Navigation with Guaranteed Feature-Observability for Visual-Inertial Odometry”

    2608.02079
  8. cs.CL arXiv submitted 2026-08-03
    Aligning evidence to boundaries lifts video grounding by ~1.9 pts

    Wei Jia +7 · “CAVE: Competence-Aware Visual Boundary Evidence Alignment for Video Temporal Grounding”

    2608.02078
  9. math.OC arXiv submitted 2026-08-03
    Observed inputs can speed evolutionary optimization

    So Nakashima +1 · “Accelerating Evolutionary Strategy via Rao-Blackwellizing Realization of Uncertain Input”

    2608.02073
  10. cs.CY arXiv submitted 2026-08-03
    LOCO 2026 proceedings archive computing's carbon-reduction research

    Adrian Friday +5 · “Proceedings of the 2nd International Workshop on Low Carbon Computing (LOCO 2026)”

    2608.02072
  11. cs.LG arXiv submitted 2026-08-03
    FFN layers steer residual directions

    Timur Mudarisov +2 · “Feed-Forward Steering in Transformer Residual Dynamics”

    2608.02071
  12. cs.CV arXiv submitted 2026-08-03
    Shared soft-MoE EEG model ranks first on 14 evaluation settings

    Zhu Chen +3 · “STEAM: A Spatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding”

    2608.02070
  13. cs.RO arXiv submitted 2026-08-03
    20-minute training yields deployable blind quadruped policies

    Martin Opat · “Open-DiffLoco: Open-Source Differentiable Learning for Deployable Blind Quadruped Locomotion”

    2608.02069
  14. cs.CV arXiv submitted 2026-08-03
    Mirror and glass depth errors cut up to 38% with no depth labels

    Xianghui Fan +8 · “GIFT: Geometry-Invariant Fine-Tuning for Non-Lambertian Monocular Depth Estimation”

    2608.02068
  15. cs.NI arXiv submitted 2026-08-03
    Finer RAN control timing lifts slice performance up to 99%

    Maria Tsampazi · “Cross-Layer Optimization and System-Level Design of Next-Generation Wireless Networks via Intelligent RAN Control”

    2608.02067
  16. cs.LG arXiv submitted 2026-08-03
    Geometry-guided feed-forward widths beat uniform and cosine

    Timur Mudarisov +2 · “Geometry-Guided Layerwise FFN Width Allocation in Transformers”

    2608.02064
  17. cs.AI arXiv submitted 2026-08-03
    Reserve-correction layer adds up to 2.7 points to radio classifier

    Linzhuo Han +2 · “Cross-Fitted Residual Utility for Primary-Preserving Cognitive Decision Correction in Automatic Modulation Classification”

    2608.02063
  18. eess.SP arXiv submitted 2026-08-03
    KAN cuts FTN bit errors 18.6x vs MLP at 10 dB

    Sude Ertan +2 · “A Comparative Analysis of MLP and Kolmogorov-Arnold Networks (KAN) for Faster-than-Nyquist (FTN) Signaling Detection”

    2608.02062
  19. cs.CV arXiv submitted 2026-08-03
    Relational span signal lifts moment retrieval by up to 42%

    Ji Huang +4 · “TBSG-Net: Temporal Bipartite Scene Graph Network for Fine-Grained Video Moment Retrieval”

    2608.02056
  20. eess.IV arXiv submitted 2026-08-03
    One neural net matches retrained models for any b-value protocol

    Andrea Brigliadori +2 · “Protocol generalisation for brain tissue microstructure estimation via hypernetwork-controlled geometric deep learning”

    2608.02053
  21. cs.LG arXiv submitted 2026-08-03
    Mobility models memorize up to 85% of training trajectories

    Anne Josiane Kouam +2 · “Secrets Everywhere: Auditing Memorization in Mobility Prediction Models”

    2608.02052
  22. cs.CL arXiv submitted 2026-08-03
    Narrow-to-wide schedule, not iteration, drives TextNCA's gains

    Avni Mittal +9 · “TextNCA: Neural Cellular Automata for Language Modeling via Hierarchical Local Attention”

    2608.02050
  23. cs.CV arXiv submitted 2026-08-03
    29% on satellite video vs 70% on natural: AI models lag 40 points

    Hongjie Zhou +7 · “RSVideo: Are Your Vision-Language Models Ready for Remote Sensing Videos?”

    2608.02039
  24. cs.LG arXiv submitted 2026-08-03
    Neural elements work when they learn energy

    Hongyue Jiang +3 · “Convex Neural Energy Elements: Monolithic Finite-Element Assembly of Geometry-Parameterized Neural Operators with Stability and Error Guarantees”

    2608.02036
  25. cs.LG arXiv submitted 2026-08-03
    Recurrent states double as attention memory

    Yixiao Qian +5 · “DART: Decoded Attention over Recurrent States for Efficient Long-Context Sequence Modeling”

    2608.02032
  26. cs.AI arXiv submitted 2026-08-03
    Hypergraph pairing finds LLM reasoning bugs in fewer rollouts

    Runchuan Zhu +6 · “HPFA: Hypergraph-Based Paired Failure Attribution for LLM Reasoning”

    2608.02026
  27. cs.AI arXiv submitted 2026-08-03
    Multi-turn chats nearly double K-12 LLM attack success

    Junyeong Park +5 · “EduZone: A Framework for Evaluating LLM Safety for K-12 Students and Teachers”

    2608.02024
  28. eess.AS arXiv submitted 2026-08-03
    One model: caption or voice clip yields full audio scenes

    Yu Zhang +5 · “SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks”

    2608.02023
  29. math.NA arXiv submitted 2026-08-03
    Computable error bound found for magnetization simulations

    Stefan Karch · “A posteriori error analysis for the second-order BDF method for the Landau-Lifshitz-Gilbert equation”

    2608.02021
  30. cs.CV arXiv submitted 2026-08-03
    Human confirmation amplifies hidden prompt injections in AI agents

    Jia-Chen Zhang +2 · “Invisible Ink Threats: Adversarial Goals Behind Legitimate Tasks in Computer-Use Agents”

    2608.02018
  31. cs.CV arXiv submitted 2026-08-03
    Chunk-wise VAE scales 3D modeling to 1536³

    Kaiyi Zhang +10 · “Beyond Global Latents: Chunk-Based Sparse Grid VAE for Scalable 3D Modeling”

    2608.02016
  32. math.NA arXiv submitted 2026-08-03
    Least-squares unfitted FEM: optimal order

    Jiashun Hu +2 · “An $H^{-1}$ least-squares UnCut FEM on domains defined by a level set function”

    2608.02015
  33. cs.RO arXiv submitted 2026-08-03
    One grasp model works on hands it never trained on

    Heng Zhang +4 · “MANGO-Grasp: Mahalanobis Fields over Geometry-Oriented 3D Gaussians for Cross-Embodiment Dexterous Grasping”

    2608.02014
  34. cs.LO arXiv submitted 2026-08-03
    New logic combines two decidable fragments and stays decidable

    Emanuel Kiero\'nski +2 · “The Triguarded Fragment”

    2608.02013
  35. cs.AI arXiv submitted 2026-08-03
    Answering before reading causes many agentic RAG errors

    Daeyoung Roh +1 · “Before Reasoning Can Fail: Pre-Evidence Procedural Failures in Agentic RAG”

    2608.02011
  36. cs.AI arXiv submitted 2026-08-03
    Stop search agents when evidence covers every needed fact

    Daeyoung Roh +1 · “HALT: Verification-Aware Stopping for Retrieval-Augmented Search Agents”

    2608.02009
  37. cs.CV arXiv submitted 2026-08-03
    Motion tracks beat pixel colors for unsynced 3D video

    Junyu Zhu +5 · “ASTRA: Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment”

    2608.02006
  38. cs.AI arXiv submitted 2026-08-03
    Gated skill updates tether LLM agents to equilibrium

    Zhaofeng Zhang +5 · “Evolving in the Agent Jungle via History-Informed Opponent Awareness”

    2608.02005
  39. cs.DL arXiv submitted 2026-08-03
    BIP! Ranker computes eight citation metrics on billion-edge graphs

    Ilias Kanellos +2 · “BIP! Ranker: A Software Library for Citation-Based Impact Indicators on Large-Scale Graphs”

    2608.02004
  40. cs.SE arXiv submitted 2026-08-03
    Best coding agent finds under 23% of hard repository vulnerabilities

    Kexing Ji +9 · “VulnGym: Benchmarking Coding Agents for Repository-Level Vulnerability Detection”

    2608.02001
  41. eess.SY arXiv submitted 2026-08-03
    Properly tuned sliding-mode observer beats I&I in replication

    A.J. Munoz-Vazquez +1 · “A Fair Comparison of Sliding-Mode and Immersion-and-Invariance Observers”

    2608.02000
  42. cs.DC arXiv submitted 2026-08-03
    Lyapunov scheduler times VLM offloads for timely driving

    Mengmeng Zhu +3 · “TALSC: Timeliness-Aware Large-Small VLM Collaboration for Infrastructure-Assisted Autonomous Driving”

    2608.01998
  43. cs.LG arXiv submitted 2026-08-03
    Optimizer switcher tops single-method training on 6 of 8 benchmarks

    Alok Kumar Pandey +3 · “AOS: Adaptive Optimizer Switching via Training-State Signals for Faster Convergence and Better Generalization”

    2608.01997
  44. cs.AI arXiv submitted 2026-08-03
    Workflow, not model, drives long-horizon AI research behavior

    Aon Safdar +1 · “Long-Horizon Autonomous Architecture Research with a Language-Model Agent: A Behavioural Case Study”

    2608.01995
  45. cs.CE arXiv submitted 2026-08-03
    Transport certificate shows switching diffusions contract exponentially

    Yutong Zhu +1 · “Mode-Weighted Transport Certificates for State-Dependent Reflected Switching Diffusions”

    2608.01992
  46. cs.CV arXiv submitted 2026-08-03
    Zero-parameter affinity loss cuts ImageNet FID in diffusion transformers

    Zong-Wei Hong +5 · “SPARE: Structural Parameter-Free Affinity Regularization for Flow Matching”

    2608.01990
  47. cs.CV arXiv submitted 2026-08-03
    AI detector names the fake pixels and explains why

    Kun Guo +5 · “Grounding and Explaining Visual Evidence for AI-Generated Image Detection in Human-Centric Scenes”

    2608.01988
  48. cs.CC arXiv submitted 2026-08-03
    Full Euclidean transcript defeats constant-depth arithmetic

    Amiel Ferman · “The Complete Extended Euclidean Scheme Is Not in Piecewise Arithmetic $\mathrm{AC}^0$”

    2608.01987
  49. cs.CE arXiv submitted 2026-08-03
    One RL policy matches reservoir history in 200 simulator calls

    Nam-Phong Huu Nguyen +2 · “Physics-Constrained Soft Actor-Critic for Simulator-in-the-Loop Petroleum Reservoir History Matching”

    2608.01986
  50. cs.CV arXiv submitted 2026-08-03
    Token pruning via noise throttling keeps 96.5% accuracy

    Landi He +3 · “DiffPrune: differentiable information throttling for token pruning in vision-language models”

    2608.01985