Large language models

Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.

26,821 stories · RSS feed

arXiv AI

Jun 24

Can Language Model Agents be Helpful Circuit Explainers in Mechanistic Interpretability?

arXiv:2606. 24026v1 Announce Type: new Abstract: Mechanistic interpretability has made substantial progress in automatically localizing circuits, but explaining what localized components do remains labor-intensive and difficult to standardize.

By Ayan Antik Khan, Harsh Kohli, Yuekun Yao, Huan Sun, Ziyu Yao

llms agents benchmarks safety

arXiv AI

Jun 24

FLUX3D: High-Fidelity 3D Gaussian Generation with Diffusion-Aligned Sparse Representation

arXiv:2606. 24874v1 Announce Type: cross Abstract: Sparse voxel representation has emerged as a scalable foundation for image-to-3D Gaussian Splatting (3DGS) generation, yet current methods struggle to preserve high-frequency visual details of input images due to two structural bottlenecks.

By Haorui Ji, Weizhe Liu, Hongdong Li, Hengkai Guo

llms rag diffusion multimodal benchmarks safety

arXiv AI

Jun 24

SURGELLM: Rethinking Multi-Task Evaluation through Task-Aware Feature Gating with Class-Balanced Normalization

arXiv:2606. 24259v1 Announce Type: cross Abstract: Fine-tuned encoders deployed across heterogeneous NLP tasks face three compounding problems: mismatched inductive biases, class-imbalance corruption of feature statistics, and no mechanism to condition attention on external lexical knowledge.

By Noor Islam S. Mohammad, Ulug Bayazit

llms nlp fine-tuning efficiency safety

arXiv AI

Jun 24

Lightweight Transformer Models for On-Device Fault Detection: A Benchmark Study on Resource-Constrained Deployment

arXiv:2606. 24173v1 Announce Type: cross Abstract: On-device fault detection enables real-time diagnostics without cloud dependency, but deploying machine learning models on resource-constrained hardware demands careful tradeoffs between accuracy, latency, and model size.

By Disha Patel

llms efficiency benchmarks

arXiv AI

Jun 24

Does Mixture-of-Experts Actually Help Inference on Consumer and Edge Hardware? An Empirical Study

arXiv:2606. 21428v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) language models are often described as ideal for resource-constrained inference.

By Alfarizy Alfarizy, Hung Truong Thanh Nguyen, Ren\'e Richard, Roozbeh Razavi-Far, Hung Cao

llms efficiency benchmarks

arXiv AI

Jun 24

PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Models

arXiv:2606. 24388v1 Announce Type: new Abstract: We introduce a large-scale, open-source dataset of pre-generated adversarial attacks for vision-language models (VLMs).

By Simone Gallivanone, Hossein Khodadadi, Mauro Dore, Mauro Medda, Nicola Franco

llms fine-tuning multimodal benchmarks safety

arXiv Machine Learning

Jun 24

Systematic Exploration of 4-Expert Heterogeneous Mixture-of-Experts via Automated Pipeline Search

arXiv:2606. 23739v1 Announce Type: new Abstract: We present an automated large-scale search pipeline for heterogeneous 4-Expert Mixture-of-Experts (MoE4) architectures within the LEMUR neural network dataset ecosystem.

By Yashkumar R Lukhi, Harsh Rameshbhai Moradiya, Radu Timofte, Dmitry Ignatov

llms safety

arXiv AI

Jun 24

Deep Learning Approaches for 3D Medical Scene Completion: From Geometric Modeling to Generative Paradigms

arXiv:2606. 24180v1 Announce Type: cross Abstract: Three-dimensional scene completion has evolved as a major problem in computer vision and robotics, and its applications are diverse, including autonomous navigation and augmented reality.

By Afifa Khaled, Said Jadid Abdulkadir, Majdy Mohamed Eltayeb Eltahir

llms agents diffusion computer-vision robotics

arXiv AI

Jun 24

ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling

arXiv:2606. 24437v1 Announce Type: new Abstract: Mixture-of-Agents (MoA) architectures improve inference-time scaling by organizing multiple LLM agents into layered reasoning pipelines.

By Heng Ping, Arijit Bhattacharjee, Peiyu Zhang, Shixuan Li, Wei Yang, Ali Jannesari, Nesreen Ahmed, Paul Bogdan

llms agents efficiency benchmarks

arXiv AI

Jun 24

CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

arXiv:2606. 24467v1 Announce Type: new Abstract: Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware.

By Xiaolin Lin, Jingcun Wang, Olga Kondrateva, Yiyu Shi, Bing Li, Grace Li Zhang

llms efficiency

arXiv AI

Jun 24

DeepBD: A Grounded Agentic Workflow for Variant Prioritization and Diagnosis of Genetic Birth Defects

arXiv:2606. 24779v1 Announce Type: cross Abstract: Birth defects are a major cause of fetal loss, neonatal morbidity and long-term disability.

By Shiyu Li, Ziqi Yan, Zhihao Wu, Jielong Lu, Weiran Liao, Jiajun Yu, Genjie Li, Zeyu Chu, Jiajun Bu, Haishuai Wang

llms rag agents benchmarks

arXiv Machine Learning

Jun 24

Uniform Sampling from High-dimensional Spectral Norm Balls

arXiv:2606. 24134v1 Announce Type: cross Abstract: Motivated by an application in machine learning optimization, this paper focuses on the challenges of sampling a matrix uniformly from the unit spectral norm ball.

By Michael R. Metel

llms

arXiv AI

Jun 24

Bayesian control for coding agents

arXiv:2606. 24453v1 Announce Type: new Abstract: Modern coding agents pair LLM generators with various tools, including cheap diagnostics and expensive verifiers.

By Theodore Papamarkou, Vladislav Smirnov, Viktor Mazanov, Artem Vazhentsev, Preslav Nakov, Timothy Baldwin, Artem Shelmanov

llms agents benchmarks

arXiv AI

Jun 24

Accuracy and Satisfaction in Multi-Turn LLM Dialogues for NFR Assessment

arXiv:2606. 24834v1 Announce Type: new Abstract: LLM-based dialogue assistants have become mainstream tools for software developers, yet current evaluation benchmarks focus exclusively on functional correctness.

By Ali Pourghasemi Fatideh, Wilder Baldwin, Maria Dhakal, Collin McMillan, Sepideh Ghanavati

llms agents benchmarks

arXiv AI

Jun 24

Topological Neural Dynamics: A Neuron-wise Framework for Sequence Modeling

arXiv:2606. 21295v2 Announce Type: replace-cross Abstract: Existing sequence models, including RNNs, LSTMs, continuous-time networks, and Transformers, share a common structural principle: layer-wise dynamics, where all neurons in the same layer co-evolve through a shared parameterized operator, leaving individual neurons no freedom to evolve independently.

By Borui Cai, Yao Zhao

llms safety

arXiv AI

Jun 24

Event-Grounded Question Answering over Long Audio via Structured Retrieval

arXiv:2602. 14612v4 Announce Type: replace-cross Abstract: Answering natural-language questions over multi-hour audio requires both event recognition and temporal grounding.

By Kartik Hegde, Arvind Krishna Sridhar, Naveen Vakada, Yinyi Guo, Erik Visser

llms rag nlp benchmarks data-engineering

arXiv Machine Learning

Jun 24

NeuroSonic: Conditional Flow Matching for EEG-to-Speech Reconstruction

arXiv:2606. 24087v1 Announce Type: new Abstract: Reconstructing continuous speech from scalp electroencephalography (EEG) remains fundamentally challenging.

By Wenhao Gao, Yifan Wang, Yijia Ma, Carl Yang, Wen Li, Chenyu You

llms diffusion benchmarks

arXiv Machine Learning

Jun 24

Polaris: A Godel Agent Framework for Small Language Models through Experience-Abstracted Policy Repair

arXiv:2603. 23129v3 Announce Type: replace Abstract: G\"odel agent realize recursive self-improvement: an agent inspects its own policy and traces and then modifies that policy in a tested loop.

By Aditya Kakade, Vivek Srivastava, Shirish Karande

llms agents benchmarks

arXiv Machine Learning

Jun 24

An LLM-based Two-Stage Transformer Framework for Cross-Domain Bearing Fault Diagnosis with Limited Data

arXiv:2606. 24459v1 Announce Type: new Abstract: Bearing fault diagnosis faces critical challenges when dataset heterogeneity, operating condition variations, and limited labeled data occur simultaneously in industrial environments.

By Jinghan Wang, Feng Cheng, Wentao Wu, Hang Li, Gaoliang Peng, Tianchen Liu

llms rag fine-tuning benchmarks safety

arXiv AI

Jun 24

Toward Self-Evolution-Ready Workflow Harnesses: A Reversible Migration Path and Convertibility Taxonomy for Expert LLM Pipelines

arXiv:2606. 24598v1 Announce Type: cross Abstract: While expert-validated "LLM + script" workflows deliver significant value, they remain static: they encode hard-won domain knowledge yet fail to adapt execution based on feedback.

By Yimo Lin, Zhen Zhang, Yibin Li

llms agents benchmarks

Get the daily briefing