arXiv AI
Jun 2
Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning
arXiv:2602. 02098v2 Announce Type: replace-cross Abstract: Multi-task reinforcement learning trains generalist policies that can execute multiple tasks.
Benchmarks and evaluation
Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv AI
Jun 2
naPINN: Noise-Adaptive Physics-Informed Neural Networks for Recovering Physics from Corrupted Measurement
arXiv:2602. 02547v2 Announce Type: replace-cross Abstract: Physics-Informed Neural Networks (PINNs) are effective methods for solving inverse problems and discovering governing equations from observational data.
arXiv AI
Jun 2
Beware of the Batch Size: Hyperparameter Bias in Evaluating LoRA
arXiv:2602. 09492v2 Announce Type: replace-cross Abstract: Low-rank adaptation (LoRA) is a standard approach for fine-tuning large language models, yet its many variants report conflicting empirical gains, often on the same benchmarks.
arXiv AI
Jun 2
DenseMLLM: Standard Multimodal LLMs for Dense Prediction
arXiv:2602. 14134v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in high-level visual understanding.
arXiv AI
Jun 2
On Imbalanced Regression with Hoeffding Trees
arXiv:2602. 22101v3 Announce Type: replace-cross Abstract: Many real-world applications generate continuous data streams for regression.
arXiv AI
Jun 2
T1: Tool-integrated Verification for Test-time Compute Scaling in Small Language Models
arXiv:2504. 04718v2 Announce Type: replace-cross Abstract: Recent studies have demonstrated that test-time compute scaling effectively improves the performance of small language models (sLMs).
arXiv AI
Jun 2
Consistent and Distinctive: LLM Benchmark Efficiency via Maximum Independent Set Prompt Selection on Similarity Graphs
arXiv:2606. 01400v1 Announce Type: cross Abstract: Evaluating large language models (LLMs) across comprehensive benchmarks is expensive and time-consuming.
arXiv AI
Jun 2
Towards Understanding Modality Interaction in Multimodal Language Models via Partial Information Decomposition
arXiv:2606. 00959v1 Announce Type: new Abstract: Understanding modality interaction in multimodal large language models (MLLMs) is central to reliable deployment.
arXiv AI
Jun 2
Enhancing BiGRU with a KAN Block for Legal Document Classification and Summarization
arXiv:2606. 00116v1 Announce Type: cross Abstract: This study introduces a novel architecture of KAN-based BiGRU model for the task of classification and summarization of legal documents in a low-resource multilingual setup.
arXiv Machine Learning
Jun 2
PFT: Phonon Fine-tuning for Machine Learned Interatomic Potentials
arXiv:2601. 07742v4 Announce Type: replace-cross Abstract: Many materials properties depend on higher-order derivatives of the potential energy surface, yet machine learned interatomic potentials (MLIPs) trained with a standard loss on energy, force, and stress errors can exhibit error in curvature, degrading the prediction of vibrational properties.
arXiv Machine Learning
Jun 2
Massive Spikes in LLMs are Bias Vectors: Mechanistic Uncovering and Spike-Free Quantization
arXiv:2606. 02288v1 Announce Type: new Abstract: Massive activation spikes in Large Language Models (LLMs) severely degrade quantization by stretching dynamic ranges.
arXiv AI
Jun 2
Predicting Future Utility: Global Combinatorial Optimization for Task-Agnostic KV Cache Eviction
arXiv:2602. 08585v2 Announce Type: replace-cross Abstract: Given the quadratic complexity of attention, KV cache eviction is vital to accelerate model inference.
arXiv Machine Learning
Jun 2
Rethinking Amortized Neural Representations for High-Resolution Terrain Elevation Data
arXiv:2606. 00404v1 Announce Type: cross Abstract: Implicit neural representations (INRs) model a signal as a continuous coordinate-to-value function.
arXiv AI
Jun 2
Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling
arXiv:2606. 02578v1 Announce Type: cross Abstract: Recent multimodal large language models have demonstrated strong reasoning ability, yet their reliability as automated evaluators remains limited by a critical weakness: when visual evidence conflicts with textual cues, MLLM judges tend to reward plausible narratives over perceptually correct answers.
arXiv AI
Jun 2
CityTrajBench: A Unified Benchmark for City-Scale Vehicle Trajectory Generation
arXiv:2606. 02287v1 Announce Type: cross Abstract: Urban trajectory generation is a fundamental task for transportation simulation, urban planning, and mobility analytics.
arXiv AI
Jun 2
Faster Synchronous On-Policy RL via Straggler-Aware Group Sizing
arXiv:2606. 02218v1 Announce Type: cross Abstract: Synchronous reinforcement learning methods such as Group Relative Policy Optimization (GRPO) provide stable and reproducible on-policy training, but they are highly vulnerable to stragglers, a single unusually long rollout can delay reward computation and parameter updates for the entire group.
arXiv AI
Jun 2
The Image Reconstruction Game: Drawing Common Ground Through Iterative Multimodal Dialogue
arXiv:2606. 01901v1 Announce Type: cross Abstract: We introduce the Image Reconstruction Game, a fully automated benchmark in which a vision-language model issues corrective instructions to an image generator across multiple turns, making accumulated common ground directly observable as a rendered image.
arXiv AI
Jun 2
DetailMaster: Can Your Text-to-Image Model Handle Long Prompts?
arXiv:2505. 16915v3 Announce Type: replace-cross Abstract: While recent Text-to-Image (T2I) models show impressive capabilities in synthesizing images from brief descriptions, they struggle with the long, detailed prompts required for professional applications.
arXiv AI
Jun 2
Boosting Multimodal Federated Learning via Chained Modality Optimization
arXiv:2606. 01856v1 Announce Type: cross Abstract: Multimodal Federated Learning (MMFL) enables privacy-preserving collaborative learning across decentralized clients with heterogeneous data and modality availability.
arXiv AI
Jun 2
Distilling Neuro-Symbolic Programs into 3D Multi-modal LLMs
arXiv:2606. 01215v1 Announce Type: cross Abstract: Current 3D spatial reasoning methods face a fundamental trade-off: neuro-symbolic 3D (NS3D) concept learners achieve interpretable reasoning through compositional programs but are constrained to closed-set concept vocabularies and simple programs; end-to-end 3D multi-modal LLMs (3D MLLMs) could handle complex natural language and open-vocabulary concepts but suffer from black-box reasoning without explicit spatial verification.