AI/ML arXiv cs.AI

CLAP: Direct VLM-to-VLA Adaptation via Language-Action Grounding

CLAP, a method for adapting VLMs to VLAs by prepending natural-language action descriptions to numeric action sequences to preserve VLM capabilities.

Software Engineering Hacker News

Converting colors in JavaScript at 6B operations per second

A discussion on achieving extremely high-performance color conversion in JavaScript, reaching 6 billion operations per second.

AI/ML arXiv cs.AI

Reward Transport: Property Control in Flow Matching via Noise-Space Alignment

Introduces Reward Transport, a method for controllable molecular generation in flow matching by aligning noise-space coordinates with rewards.

AI/ML arXiv cs.AI

Director: Accelerating Distributed MoE Serving via Online Proactive Expert Placement

Presents Director, a distributed MoE serving system that uses online proactive expert placement to reduce end-to-end latency.

AI/ML arXiv cs.AI

LieBN: Batch Normalization over Lie Groups

Proposes LieBN, a Riemannian Batch Normalization framework specifically designed for Lie groups across various geometries.

AI/ML arXiv cs.AI

HERO: A Heterogeneity-Aware Benchmark Library for Federated Continual Learning

Introduces HERO, a benchmark library for Federated Continual Learning to support reproducible and setting-aware evaluation.

AI/ML arXiv cs.AI

DaDaDa: A Dataset for Data Pricing in Data Marketplaces

Presents DaDaDa, a dataset for data product pricing in data marketplaces to establish standardized benchmarks.

AI/ML arXiv cs.AI

Accelerating GPU Inference of Large Language Models with Moderately Unstructured Sparse Weight Matrices

A new GPU inference method for LLMs using moderately unstructured sparse weight matrices to outperform dense matrix multiplication.

AI/ML arXiv cs.AI

LLM-Driven Evolutionary Generation of Multi-Objective Bayesian Optimization Algorithms

Uses LLMs to evolutionarily generate multi-objective Bayesian optimization algorithms, achieving high efficiency and accuracy.

AI/ML arXiv cs.AI

EHR-MPC: Inference-Time Control for Sepsis Treatment with Generative Patient Digital Twins

EHR-MPC decouples patient dynamics learning from treatment optimization using generative digital twins for sepsis treatment.

AI/ML arXiv cs.AI

Multi-Conditioned Diffusion Synthesis of Sand Boils for Low-Resource Earthen-Levee Inspection

A diffusion-based synthesis pipeline for generating synthetic sand-boil imagery to improve earthen-levee inspection.

Hardware/Chips Hacker News

Beavis Ultrasound PnP ISA Sound Card Replica

A replica of the Beavis Ultrasound PnP ISA Sound Card, recreating vintage computer hardware.

AI/ML arXiv cs.AI

TrustX Agent Risk Classification Framework (ARC): Risk-Tiering Internally Created Agentic AI Systems

The TrustX Agent Risk Classification Framework (ARC) provides a structured method for risk-tiering and governing agentic AI systems.

AI/ML arXiv cs.AI

Agora: Enhancing LLM Agent Reasoning Via Auction-Based Task Allocation

Agora is a framework that uses an auction-based mechanism to dynamically allocate tasks to expert LLM models and tools based on competence.

AI/ML arXiv cs.AI

ConceptSMILE: Auditing the Trustworthiness of Concept-Based Explainable AI

ConceptSMILE is a perturbation-based auditing framework designed to evaluate the reliability of concept-based explainable AI (XAI).

AI/ML arXiv cs.AI

Minimal Decision Dynamics and Contextual Probability: A Quantum Tug-of-War Model

A research paper proposing a quantum-like extension of the Tug-of-War model to represent contextual probability in decision-making dynamics.

AI/ML arXiv cs.AI

REFORGE: A Method for Benchmarking LLMs' Reverse Engineering Capabilities in Decompiled Binary Function Naming

Reforge is a provenance-tracked pipeline for benchmarking the reverse engineering capabilities of LLMs in decompiled binary function naming.

AI/ML arXiv cs.AI

A Unified Approach to Interpreting Knowledge Distillation for Large Language Models via Interactions

A new approach to knowledge distillation in LLMs using 'interactions' and a 'Complex Interaction Penalty' (CIP) loss function to improve performance.

AI/ML arXiv cs.AI

iLENS: Interpretable LLM-Guided Mixture-of-Experts for Neuroimaging Survival Analysis

iLENS is an interpretable LLM-guided Mixture-of-Experts framework for neuroimaging survival analysis in Alzheimer's Disease prediction.

AI/ML arXiv cs.AI

Signed Symmetric Quantization for Few-Bit Integers

Signed Symmetric Quantization is proposed as a lightweight alternative to asymmetric quantization for few-bit integers in LLMs, reducing error without runtime penalties.