archive
Every paper Pith has read. Search by title, abstract, or pith.
14513 papers in cs.AI · page 5
-
Compiler turns programs into exact neural modules
The Neural Compiler: Program-to-Network Translation for Hybrid Scientific Machine Learning
-
Multimodal policies fail differently depending on latent or generative setup
Understanding Multimodal Failure in Action-Chunking Behavioral Cloning
-
Random node sampling matches full GNN training on most datasets
Implicit Regularization of Mini-Batch Training in Graph Neural Networks
-
Spectral alignment boosts cross-subject F1 in biomedical signals
BioFormer: Rethinking Cross-Subject Generalization via Spectral Structural Alignment in Biomedical Time-Series
-
Method finds selective features only partially causal for IOI task
From Correlation to Cause: A Five-Stage Methodology for Feature Analysis in Transformer Language Models
-
Knowledge graph becomes factory's real-time authoritative state
KAPPS: A knowledge-based CPPS Architecture for the Circular Factory
-
LLM driving planner cuts lag from 3s to near zero
Steins;Gate Drive: Semantic Safety Arbitration over Structured Futures for Latency-Decoupled LLM Planning
-
Synthetic RAW data yields same low-light detection metrics as real
Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light
-
Rehearsal of stored Q-values cuts forgetting in repeated RL tasks
Don't Forget the Critic: Value-Based Data Rehearsal for Multi-Cyclic Continual Reinforcement Learning
-
S2ED keeps story characters consistent across frames
S2ED: From Story to Executable Descriptions for Consistency-Aware Story Illustration
-
Pre-VLA lifts VLA success rates from 31% to 38%
Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts
-
Constant-time activations eliminate timing leaks on microcontrollers
A Constant-Time Implementation Methodology for Activation Functions on Microcontrollers
-
EM pulses trigger persistent accuracy collapse on NCS2 until reload
Characterizing the Fault Response of the Intel Neural Compute Stick 2 Under Single-Pulse Electromagnetic Fault Injection
-
Table structure recovered by predicting grid counts and separators directly
FastTab: A Fast Table Recognizer with a Tiny Recursive Module and 1D Transformers
-
GenRe generalizes 3D urban scenes to new viewpoints in minutes
Diffusion-guided Generalizable Enhancer for Urban Scene Reconstruction
-
Lesion grounding lifts ophthalmic VQA accuracy and clarity
Towards Clinically Interpretable Ophthalmic VQA via Spatially-Grounded Lesion Evidence
-
Query-time RL turns noisy memory into accurate evidence
DeferMem: Query-Time Evidence Distillation via Reinforcement Learning for Long-Term Memory QA
-
Three models embed ingredients via recipe and chemistry graphs
Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
-
Local EEG segment matching raises cross-subject emotion accuracy
Cross-Subject EEG Emotion Recognition Based on Temporal Asynchronous Alignment Contrastive Learning
-
Adversarial scaling reveals LLM code weaknesses
VeriScale: Adversarial Test-Suite Scaling for Verifiable Code Generation
-
TimeGuard boosts backdoor resistance in time series forecasts by 1.96x
TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting
-
POMDP decomposition scales sensor selection by three orders
Scaling Observation-aware Planning in Uncertain Domains
-
Nash bargaining in RL lifts EV energy trading welfare 61 percent
Incentive-Aligned Vehicle-to-Vehicle Energy Trading via Nash-Integrated Multi-Agent Reinforcement Learning
-
Multiple metrics needed to judge liver vessel segmentation
VEELA: A Clinically-Constrained Benchmark for Liver Vessel Segmentation in Computed Tomography Angiography
-
LLMs learn to plan transit routes from records alone
TransitLM: A Large-Scale Dataset and Benchmark for Map-Free Transit Route Generation
-
MLLM planner in ViT space guides DiT to SOTA video generation and edits
Bernini: Latent Semantic Planning for Video Diffusion
-
Trial harnesses let AI agents turn outcomes into process updates
Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators
-
Watermarking 4D splats by gating at motion-curvature instants
4D-GSW: Kinematic-Aware Spatio-Temporal Consistent Watermarking for 4D Gaussian Splatting
-
Dynamic meta-tokens compress KV cache while preserving context
Meta-Soft: Leveraging Composable Meta-Tokens for Context-Preserving KV Cache Compression
-
Scaling sepsis AI replicas to CPU threads halves detection latency
SepsisAI Orchestrator: A Containerized and Scalable Platform for Deploying AI Models and Real-Time Monitoring in Early Sepsis Detection
-
Attacks lift autonomous agent risk rate from 28.3% to 52.6%
Benchmarking Autonomous Agents against Temporal, Spatial, and Semantic Evasions
-
Deep RL agent outperforms rules at O-RAN conflict resolution
ACCoRD: Actor-Critic Conflict Resolution with Deep learning for O-RAN xApps
-
Benchmark compares 12 pipelines for knowledge graph integration
Evaluation of Pipelines for Data Integration into Knowledge Graphs
-
Benchmarks reveal when AI coordination improves scientific inference
Cross-domain benchmarks reveal when coordinated AI agents improve scientific inference from partial evidence
-
Heavy-tailed spectra set per-layer rates to speed LLM training 1.5x
One LR Doesn't Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs
-
Three modules let 8B model rival big LLMs on chemistry
SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules
-
EmoTrack cuts depression prediction error 13.5% on single sessions
EmoTrack: Robust Depression Tracking from Counseling Transcripts across Session Regimes
-
Multi-grained compression lifts long video QA accuracy
MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering
-
YOLOv8 recall falls below 40% under strong turbulence in satellite images
Impact of Atmospheric Turbulence and Pointing Error on Earth Observation
-
Probe set partitions flag atypical federated clients
Detecting Atypical Clients in Federated Learning via Representation-Level Divergence
-
Adaptive distillation preserves exploration for better LLM math scores
Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning
-
Rewriting equivalents at test time stabilizes LLM theorem proving
What are the Right Symmetries for Formal Theorem Proving?
-
AI recommender lifts Cox fall-risk model C-index from 0.805 to 0.815
Explainable AI for Data-Driven Design of High-Dimensional Predictive Studies
-
Latent user concerns turn passive dialogue agents proactive
Unlocking Proactivity in Task-Oriented Dialogue
-
Gemini wins 20 of 32 timed Risk games but planners tie with fixed executor
Evaluating Large Language Models as Live Strategic Agents: Provider Performance, Hybrid Decomposition, and Operational Gaps in Timed Risk Play
-
Attention mask forces transformer backtracking to ignore search history
Can Transformers Learn to Verify During Backtracking Search?
-
Agents reach sites but set wrong filters on data tasks
SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval
-
Subjective logic turns assurance arguments into confidence networks
Towards a compositional semantics for quantitative confidence assessment in assurance arguments
-
Editing correct traces cuts repetition in LLM reasoning
CLORE: Content-Level Optimization for Reasoning Efficiency
-
Spike timing distinguishes objects whose features are merely rearranged
Temporal Coding as a Substrate for Sensorimotor Object Inference: A Spiking Reinterpretation of Thousand Brains Architecture