3/10/2026
Photons = Tokens: The Physics of AI and the Economics of Knowledge
Quantitative analysis of AI computation costs using thermodynamic principles, defining tokens as measurable physical quantities.
3/10/2026
SmartBench: Evaluating LLMs in Smart Homes with Anomalous Device States and Behavioral Contexts
SmartBench evaluates LLMs on anomalous device detection and reasoning in smart home environments beyond basic command execution.
3/10/2026
HEARTS: Benchmarking LLM Reasoning on Health Time Series
HEARTS benchmark evaluates LLM reasoning on diverse health time series with complex temporal dependencies across multiple physiological modalities.
3/10/2026
RECAP: Local Hebbian Prototype Learning as a Self-Organizing Readout for Reservoir Dynamics
RECAP applies Hebbian learning and local plasticity to reservoir computing for image recognition without backpropagation.
3/10/2026
SR-TTT: Surprisal-Aware Residual Test-Time Training
SR-TTT improves test-time training language models by using surprisal-aware residual learning to fix catastrophic failures on exact-recall tasks.
3/10/2026
Trust Aware Federated Learning for Secure Bone Healing Stage Interpretation in e-Health
Federated learning framework for medical imaging using trust-aware mechanisms to handle unreliable participants in distributed environments.
3/10/2026
Performance Comparison of IBN orchestration using LLM and SLMs
Compares LLMs and SLMs for Intent-Based Networking orchestration in 5G/6G using hierarchical multi-agent architecture.
3/10/2026
ObjChangeVR: Object State Change Reasoning from Continuous Egocentric Views in VR Environments
ObjChangeVR uses multimodal LLMs to reason about object state changes from continuous egocentric VR views with sparse motion cues.
3/10/2026
Geodesic Gradient Descent: A Generic and Learning-rate-free Optimizer on Objective Function-induced Manifolds
Geodesic gradient descent optimizer for learning on objective function-induced manifolds without manual learning rate tuning.
3/10/2026
PaLMR: Towards Faithful Visual Reasoning via Multimodal Process Alignment
PaLMR framework aligns multimodal LLM reasoning at both process and outcome levels using reinforcement learning to reduce process hallucinations.
3/10/2026
A Parameter-efficient Convolutional Approach for Weed Detection in Multispectral Aerial Imagery
FCBNet, parameter-efficient CNN for weed segmentation in multispectral aerial imagery using frozen ConvNeXt backbone and feature correction blocks.
3/10/2026
GameVerse: Can Vision-Language Models Learn from Video-based Reflection?
GameVerse benchmark evaluates vision-language models using reflect-and-retry paradigm in video game environments to assess learning from failure.
3/10/2026
Graph-of-Mark: Promote Spatial Reasoning in Multimodal Language Models with Graph-Based Visual Prompting
Graph-based visual prompting technique for multimodal language models that improves spatial reasoning by representing marked objects as graph structures.
3/10/2026
Accelerating Video Generation Inference with Sequential-Parallel 3D Positional Encoding Using a Global Time Index
Optimizes Diffusion Transformer video generation inference using sequential-parallel 3D positional encoding to reduce memory and latency bottlenecks.
3/10/2026
Better Eyes, Better Thoughts: Why Vision Chain-of-Thought Fails in Medicine
Analysis showing chain-of-thought prompting underperforms in medical vision tasks due to perception bottlenecks in VLMs.
3/10/2026
Hybrid Orchestration of Edge AI and Microservices via Graph-based Self-Imitation Learning
Graph-based self-imitation learning for orchestrating edge AI and microservices with latency constraints.
3/10/2026
Chart Deep Research in LVLMs via Parallel Relative Policy Optimization
Parallel Relative Policy Optimization for improving chart understanding and deep research capabilities in large vision-language models.
3/10/2026
Thinking with Gaze: Sequential Eye-Tracking as Visual Reasoning Supervision for Medical VLMs
Eye-tracking gaze trajectories used as supervision to improve vision-language model reasoning in medical radiology tasks.
3/10/2026
Mining Beyond the Bools: Learning Data Transformations and Temporal Specifications
Method for mining temporal specifications and data transformations from execution traces using syntax-guided synthesis.
3/10/2026
Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces
ATLAS framework for reinforcement finetuning enables small language models to operate effectively over large tool ecosystems with long-horizon planning and sparse rewards.
3/10/2026
ProtAlign: Contrastive learning paradigm for Sequence and structure alignment
ProtAlign: Contrastive learning framework aligning protein sequence and structure embeddings for improved protein language models.
3/10/2026
Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting
Hybrid approach combining time series foundation models with regression for electricity price forecasting capturing temporal and cross-variate patterns.
3/10/2026
Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment
Safe Transformer: Modular safety alignment via explicit discrete safety bit bottleneck for interpretable and controllable LLM behavior.
3/10/2026
Don't Freeze, Don't Crash: Extending the Safe Operating Range of Neural Navigation in Dense Crowds
Reinforcement learning for safe navigation through variable-density crowds generalizing beyond training densities without freezing.
3/10/2026
Agent Hunt: Bounty Based Collaborative Autoformalization With LLM Agents
Agent Hunt: Multi-agent autoformalization system using bounty-based marketplace where LLM agents propose and prove lemmas in theorem proving.
3/10/2026
Rank-Factorized Implicit Neural Bias: Scaling Super-Resolution Transformer with FlashAttention
Super-Resolution Transformer optimization removing relative positional bias constraints to enable FlashAttention and efficient inference.
3/10/2026
ResearchEnvBench: Benchmarking Agents on Environment Synthesis for Research Code Execution
ResearchEnvBench: Benchmark evaluating autonomous agents on environment synthesis and complex software dependency resolution for research code execution.
3/10/2026
ViroGym: Realistic Large-Scale Benchmarks for Evaluating Viral Proteins
ViroGym: Benchmark for evaluating protein language models on viral protein variant effects prediction with in silico and in vitro validation.
3/10/2026
Heterogeneous Decentralized Diffusion Models
Heterogeneous decentralized diffusion training framework reducing computational requirements while supporting diverse training objectives across distributed experts.
3/10/2026
Improved Constrained Generation by Bridging Pretrained Generative Models
Framework for constrained generation bridging pretrained models to respect complex physical constraints in robotic control and autonomous driving.
3/10/2026
Stabilizing Reinforcement Learning for Diffusion Language Models
Method stabilizing GRPO reinforcement learning for diffusion language models by addressing reward collapse and incompatibilities with dLLMs.
3/10/2026
Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection
DIRECTER: Dynamic rejection steering method to enhance LLM instruction following while avoiding oversteering degradation via activation steering.
3/10/2026
ButterflyViT: 354$\times$ Expert Compression for Edge Vision Transformers
ButterflyViT: Expert compression technique achieving 354× compression for sparse Mixture of Experts Vision Transformers on edge devices.
3/10/2026
Property-driven Protein Inverse Folding With Multi-Objective Preference Alignment
Property-driven protein inverse folding with multi-objective preference alignment for sequence design balancing designability and developability.
3/10/2026
Robotic Foundation Models for Industrial Control: A Comprehensive Survey and Readiness Assessment Framework
Survey of robotic foundation models for industrial control, analyzing applicability, requirements, and use cases in collaborative robotics.
3/10/2026
Failure Detection in Chemical Processes using Symbolic Machine Learning: A Case Study on Ethylene Oxidation
Symbolic machine learning approach for failure detection in chemical processes with interpretability focus.
3/10/2026
HGT-Scheduler: Deep Reinforcement Learning for the Job Shop Scheduling Problem via Heterogeneous Graph Transformers
Deep reinforcement learning with heterogeneous graph transformers for job shop scheduling.
3/10/2026
xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truth
Python package providing reusable infrastructure for evaluating attribution methods on synthetic time series data.
3/10/2026
Optimistic Policy Regularization
Mechanism for deep reinforcement learning that preserves exploratory behavior via historical trajectory buffering.
3/10/2026
"Dark Triad" Model Organisms of Misalignment: Narrow Fine-Tuning Mirrors Human Antisocial Behavior
Research on mechanistic understanding of LLM misalignment behaviors including deception and manipulation.
3/10/2026
Step-Level Visual Grounding Faithfulness Predicts Out-of-Distribution Generalization in Long-Horizon Vision-Language Models
Study on how temporal visual grounding in vision-language models affects out-of-distribution generalization.
3/10/2026
Twitch: Learning Abstractions for Equational Theorem Proving
Tool for discovering term abstractions in automated reasoning using pattern mining from proofs.
3/10/2026
Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering
Mechanistic interpretability study identifying audio-specialist attention heads in audio-language models to improve audio grounding over text dominance.
3/10/2026
Contextual Counterfactual Credit Assignment for Multi-Agent Reinforcement Learning in LLM Collaboration
C3 method for credit assignment in multi-agent RL systems with LLMs using counterfactual reasoning to improve learning from sparse terminal feedback.
3/10/2026
Supporting Artifact Evaluation with LLMs: A Study with Published Security Research Papers
Study using LLMs to automate artifact evaluation for published security research papers, improving reproducibility checking scalability.
3/10/2026
A prior information informed learning architecture for flying trajectory prediction
Dual-Transformer-Cascade framework for trajectory prediction integrating environmental priors for flying objects with hardware efficiency considerations.
3/10/2026
Physics-informed AI Accelerated Retention Analysis of Ferroelectric Vertical NAND: From Day-Scale TCAD to Second-Scale Surrogate Model
Physics-informed surrogate model for ferroelectric vertical NAND retention analysis, accelerating simulations from day-scale TCAD to second-scale predictions.
3/10/2026
MindfulAgents: Personalizing Mindfulness Meditation via an Expert-Aligned Multi-Agent System
Multi-agent LLM system personalizing mindfulness meditation experiences through expert-aligned design to improve user engagement.
3/10/2026
How Private Are DNA Embeddings? Inverting Foundation Model Representations of Genomic Sequences
Privacy analysis demonstrating inversion attacks on DNA embeddings from foundation models shared via Embeddings-as-a-Service frameworks.
3/10/2026
Post-Training with Policy Gradients: Optimality and the Base Model Barrier
Theoretical analysis of policy gradient post-training for linear autoregressive models with outcome and process rewards on margin-separated sequences.