INDEX
Every paper Pith has read and judged, in one searchable index.
-
cs.CV arXiv submitted 2026-07-30Monocular 3D Gaussians get language at one-tenth the memory
Pouya Ardekhani +3 · “MonoVoc: Decoupling Geometry and Semantics for Lightweight Monocular Open-Vocabulary 3D Gaussians”
2607.28300 -
cs.DS arXiv submitted 2026-07-30Max-circularity regions are hard yet pseudopolynomial
Jan-Henrik Haunert +3 · “Finding Regions of Maximum Circularity in Plane Geometric Graphs”
2607.28298 -
cs.CV arXiv submitted 2026-07-30Depth lets tiny grasp networks hit the accuracy–size Pareto front
Edoardo Ragusa +4 · “Filling the Pareto-Optimal Front for Affordance Segmentation on Embedded Devices Using RGB-D Cameras”
2607.28293 -
cs.CL arXiv submitted 2026-07-304-bit LLM fact edits stay stable with degradation debt
Anubhav Lakra +1 · “CACHE-UK: A Stability-Aware Memory Editor for Sequentially Updated Quantized LLMs in Finance”
2607.28292 -
cs.DC arXiv submitted 2026-07-30Compute schedulers are policy—and federation can act like bad traffic
Konstantinos (Kostas) E. Zachariadis +4 · “FAIR-Compute: A Roadmap for Fair and Efficient Allocation of Federated Digital Research Infrastructure”
2607.28290 -
cs.AI arXiv submitted 2026-07-30When to model a game beats perfect replay accuracy
Jens Lehmann +2 · “Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3”
2607.28287 -
math.NA arXiv submitted 2026-07-30Two fewer matrix multiplies for high-degree polynomials
J. Ib\'a\~nez +3 · “A MATLAB Tool for the Stable Generation of Matrix Polynomial Evaluation Schemes with Two-Product Savings”
2607.28286 -
cs.CL arXiv submitted 2026-07-30Multi-LLM Elo rankings match expert text judgments
Bertil Braun +1 · “(Towards) Scalable Reliable Automated Evaluation with Large Language Models”
2607.28282 -
cs.IT arXiv submitted 2026-07-30First exact capacities for nontrivial sticky channels
Mladen Kova\v{c}evi\'{c} · “The Capacity of a Family of Sticky Channels”
2607.28281 -
eess.SY arXiv submitted 2026-07-30PCM storage cuts heating peaks 5–11% when control tracks demand
Xin Jin +3 · “Adaptive Demand-Driven Energy Management of PCM-Integrated District Heating Systems: Operational Flexibility and Techno-Economic Assessment”
2607.28280 -
cs.CV arXiv submitted 2026-07-30Multi-scale CNN plus Mamba tops HSI benchmarks at 0.2M params
Jianjun Chen +6 · “MSCM-net: A hyperspectral image classiffcation method based on multi-scale convolution and Mamba”
2607.28277 -
cs.IT arXiv submitted 2026-07-30Query-steered semantic codec beats baselines on unseen objects
Sin-Yu Huang +1 · “Generalized Query-Oriented Image Semantic Coding Empowered by Large AI Models and Semantic-Aware Hybrid Beamforming”
2607.28276 -
cs.CL arXiv submitted 2026-07-30Greek morphology benchmark separates rules from memorized forms
Ioakeim Perros +3 · “MORFES: A Benchmark for Productive Inflectional Competence in Modern Greek”
2607.28274 -
cs.FL arXiv submitted 2026-07-30Linear complexity caps distinct factor frequencies
Jason Bell +2 · “Frequencies of subwords in words of linear subword complexity”
2607.28273 -
cs.AI arXiv submitted 2026-07-30Agents win by rewriting memory
Rong Wu +12 · “MemHarness: Memory Is Reconstructed, Not Replayed”
2607.28272 -
cs.SE arXiv submitted 2026-07-30Locksmith loop pushes COBOL–Java tests past search plateaus
Andras Ferenczi +4 · “Agentic Method for Deterministic Validation of Legacy Code Migration”
2607.28271 -
eess.SY arXiv submitted 2026-07-30DC-link control that respects real source limits keeps converters online
Carlo de Paolis Robles +3 · “Grid-Forming Converter DC-link Control Considering the Primary Energy Source”
2607.28270 -
cs.CV arXiv submitted 2026-07-30Blind LLM judge scores 200k disaster captions for distillation
Simone Giano +3 · “Theia: Large-Scale Multimodal Captioning and Automated Validation of the Incidents1M Dataset for Data-Free Distillation”
2607.28269 -
cs.AI arXiv submitted 2026-07-30LLM rewrites double constraint solve speed on held-out tests
Kostis Michailidis +3 · “LLM-Guided Evolutionary Search for Constraint Model Reformulation to Improve Solver Efficiency”
2607.28268 -
cs.CL arXiv submitted 2026-07-30Mid-layer caches make long reads cost fixed
Hanzuo Liu +8 · “Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory”
2607.28263 -
cs.CV arXiv submitted 2026-07-30High-noise stages set the camera
Jiwen Liu +7 · “TARS: Timestep-Aware Data Scaling for 3D-Free Video Re-Shooting”
2607.28261 -
quant-ph arXiv submitted 2026-07-30Sparse data loads with optimal square-root T-count
Tongyang Li +5 · “Optimal T Counts under Sparsity: from QROM to State Preparation and Block Encoding”
2607.28260 -
cs.LG arXiv submitted 2026-07-30Graphs become short topological tokens Transformers can read
Md Joshem Uddin +3 · “TopoFormer: Topology Meets Attention for Graph Learning”
2607.28259 -
cs.AI arXiv submitted 2026-07-30Better placement candidates lift pallet fill to 0.49
Dheeraj Poolavaram +2 · “Operationally Guided Placement-Aware Learning for Industrial Online 3D Bin Packing”
2607.28257 -
cs.RO arXiv submitted 2026-07-30Robots ship 43× smaller image tokens
Peizheng Li +3 · “When Robots Exchange Meaning: A Demo of Goal-Oriented Semantic Communications for Collaborative Robotics”
2607.28256 -
eess.SY arXiv submitted 2026-07-30Failure predictor steers robot data
Linxuan He +7 · “Self-Evolving Learning for Embodied AI with Criticality Model”
2607.28251 -
stat.ML arXiv submitted 2026-07-30Multi-basin ensembles still set the UQ standard
H. Martin Gillis +1 · “Uncertainty quantification for trustworthy deep learning: Methods and measures”
2607.28248 -
cs.CV arXiv submitted 2026-07-30Street photos lift crop maps when fused with satellites
Iason Tsardanidis +4 · “Space2Ground 2.0: A Multi-Source Dataset and Framework for Agricultural Monitoring through Fusion of Street-Level and Satellite Imagery”
2607.28247 -
cs.CV arXiv submitted 2026-07-30Synthetic ego videos lift real robot OOD success to 84% and 70%
Zexuan Yan +11 · “EgoGenesis: Egocentric World-Action Modeling with Online Anchored Projective Memory and Action-3D RoPE”
2607.28243 -
cs.SE arXiv submitted 2026-07-30Agents become cloud services for the metaverse
Xiaofei Xu +5 · “Agentic Metaverse Services: A New As-a-Service Paradigm”
2607.28242 -
cs.DS arXiv submitted 2026-07-30Consistent subsets solved in 3^{c·tw} n^{tw} time
Robert Ganian +2 · “Improved Learning with Structure: Fine-Grained Complexity of Minimum Consistent Subset”
2607.28240 -
cs.HC arXiv submitted 2026-07-30AI levels up counterspeech by making replies more elaborate
Han Li +3 · “Identifying a Level-up Pathway for AI-assisted Counterspeech through Elaboration”
2607.28239 -
cs.AI arXiv submitted 2026-07-30AI helps basic Islamic learning
Muhammad Sajjad Akbar · “AI and Authenticity in Islamic Research: A Critical Evaluation of Generative AI Reliability, Hallucination, and Source Fidelity in Quranic, Hadith, and Fiqh Knowledge”
2607.28237 -
cs.AI arXiv submitted 2026-07-30Tiny autoencoder steadies BERT under word noise
Sina Heydari +3 · “CDAE: Enhancing Perturbation Robustness in Pretrained Language Models with Contrastive Denoising”
2607.28236 -
math.NA arXiv submitted 2026-07-30High-order coarse polynomials keep multigrid CG stable
Davide Polverino +2 · “R3MG-C: a high-order algebraic-geometric multilevel preconditioner for continuous finite element discretizations”
2607.28235 -
cs.AR arXiv submitted 2026-07-30Why real DRAM chips flip both ways under RowHammer
Haocong Luo +7 · “Demystifying DRAM Read Disturbance: Bridging the Gap Between Experimental Characterization and Device-Level Modeling of RowHammer and RowPress Phenomena”
2607.28233 -
cs.CL arXiv submitted 2026-07-30One literature layer lifts life-science AI agents across four tasks
Luigi Sigillo +11 · “EMBL AI Librarian: Life-Sciences Knowledge Layer for AI Agents”
2607.28229 -
cs.AI arXiv submitted 2026-07-30Real-phone GUI agent hits 92% on live mobile tasks
Hanzhang Zhou +15 · “Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI Agents”
2607.28227 -
cs.CR arXiv submitted 2026-07-30Robot world models turn digital hacks into physical risk
Fazhong Liu +5 · “Security of World-Model-Based Embodied AI: A Lifecycle of Threats, Defenses, and Evaluation”
2607.28226 -
cs.CV arXiv submitted 2026-07-30Self-judging agents stop faking helpful image crops
Haoqing Wang +4 · “FaithEyes: Towards Faithful Tool Use via Multi-Agent Process-Image Verification”
2607.28225 -
cs.DC arXiv submitted 2026-07-30Pending GPU jobs split into quotable and never-admit
Sohan Kunkerkar · “Queue-Theoretic Admission Control for Multi-Tenant GPU Clusters”
2607.28223 -
physics.ao-ph arXiv submitted 2026-07-30AI weather models match physics at 10–15 days but blur heat extremes
Cas Decancq +3 · “Weather Emulators at the Frontier of Heat Extremes Predictability”
2607.28220 -
q-bio.PE arXiv submitted 2026-07-30Grid size flips evolution from tiny replicators to giants
Ilya Horiguchi +1 · “Hash Chemistry: Minimal Models for Evolutionary Growth of Complexity”
2607.28219 -
cs.NI arXiv submitted 2026-07-30Unpredictable RTT cuts remote-desktop activity harder than delay
Keisuke Ishibashi +6 · “Observing the Relationship between QoS Unpredictability, Prediction Error, and User Activity in a Remote Desktop Service”
2607.28216 -
cs.NI arXiv submitted 2026-07-30Wi-Fi at 6 GHz stays outside the 5G harm zone
Rafik Zitouni +7 · “Coexistence of 5G NR and Wi Fi 6E/7 at 6 GHz: Experimental Interference Measurements”
2607.28213 -
cs.CL arXiv submitted 2026-07-30Inverted sparse attention recovers causal graphs in time series
Yusen Liu +5 · “Causal Discovery with Inverted Self-attention for Multivariate Time Series”
2607.28212 -
cs.CV arXiv submitted 2026-07-30Bigger VLMs do not fix multi-attribute bias
Ioannis Sarridis +2 · “Scaling Vision-Language Models Is Not Enough to Mitigate Bias”
2607.28211 -
cs.HC arXiv submitted 2026-07-30Group EEG tracks arousal change rate
Guandong Pan +6 · “Toward Annotation-Efficient Continuous Emotion Arousal Quantification via Group-Level EEG Dynamic Neural Synchrony”
2607.28204 -
cs.HC arXiv submitted 2026-07-30Students like short AI videos
Esse Ciego +3 · “Student Perceptions and Preferences Regarding AI-Generated Instructional Videos in Computing Education”
2607.28203 -
physics.app-ph arXiv submitted 2026-07-30Agents hit 99% success on FDTR analysis from plain language
Fuwei Yang +2 · “Vibe-FDTR: An agent-oriented framework for reproducible frequency-domain thermoreflectance data analysis”
2607.28200