HN kilincarslan 9d ago

We sell digital assets built on AI-powered business models

Company marketplace selling pre-built digital assets with AI-powered business models for transfer.

HN kozika 9d ago

Recordly: A cross-platform OSS alternative to Screen Studio

Open-source cross-platform screen recorder and editor for creating product demos and walkthroughs.

HN ingve 9d ago

Your Vibe coded website is going to get you fined

Web design agency compares vibe-coded vs professionally designed websites to highlight business risks.

HN malgamves 9d ago

Camox: The framework for agent-driven websites

Camox: MIT-licensed open-source page builder framework for agent-driven websites using Claude Code.

HN Robert_Linz 9d ago

TS Foundation Models

TiRex-2 pretrained time series foundation model for zero-shot multivariate forecasting with streaming support.

HN dexmac221 9d ago

The Day I Played Hide-and-Seek with an AI

Personal narrative about playing hide-and-seek with AI-powered cameras in home.

HN dexmac221 9d ago

Running 26B and 35B LLMs at Full Speed on €990 of Used Hardware – No Cloud

Running 26B-35B LLMs locally at full speed on €990 used hardware without cloud services.

HN aborovykh 9d ago

The Data Recipe for Teaching AI New Skills [video]

Video on data recipes for teaching AI new skills.

HN javatuts 9d ago

Obsidian AI Setup – A production-ready onboarding skill for AI agents

Open-source onboarding system that sets up personalized Obsidian vaults for AI agent knowledge management.

HN mpfect 9d ago

Google's AI buildout drove 37% increase in electricity use in 2025

Google reports 37% electricity consumption increase in 2025 driven by AI data center expansion.

Ax Ye Liu, Srijan Bansal, Bo Pang, Yang Li, Zeyu Leo Liu, Yifei Ming, Zixuan Ke, Shafiq Joty, Semih Yavuz 9d ago

Procedural Memory Distillation: Online Reflection for Self-Improving Language Models

Procedural Memory Distillation method for self-improving language models that retains cross-episode procedural information from rollouts beyond episode-local rewards.

Ax Jiacheng Miao, Jonathan K Pritchard, James Zou 9d ago

The Agentic Garden of Forking Paths

Study showing AI agents capture analytical variation among human researchers; different personas enable agents to report divergent conclusions from identical datasets.

Ax Natalie Grace Brigham, Eugene Bagdasarian, Tadayoshi Kohno, Franziska Roesner 9d ago

Janus: a Playground for User-Involved Agentic Permission Management

Janus is a playground system for designing and evaluating user-involved permission management in autonomous AI agents that execute tool calls.

Ax Hongyang He, Jiuming Liu, Victor Sanchez 9d ago

Revisiting Chain-of-Thought Reasoning under Limited Supervision: Semi-supervised Chain-of-Thought Learning

Semi-CoT framework reuses LLM-generated reasoning traces as semi-supervised learning signals to improve chain-of-thought reasoning with limited labeled data.

Ax David Courtis, Wenhao Li, Scott Sanner 9d ago

OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration

OPINE-World enables data-efficient agent adaptation by synthesizing programmatic world models through LLM-generated source code refined via counterexample-guided inductive synthesis.

Ax Oskar J. Hollinsworth, Ann-Kathrin Dombrowski, Sam Adam-Day, Adam Gleave, Chris Cundy 9d ago

Scaling Trends for Lie Detector Oversight in Preference Learning

Scaling analysis of SOLiD, a lie detector-based oversight method for identifying deceptive LLM responses, showing improved scaling with larger models in preference learning.

Ax Mahyar Ghazanfari, Amin Tabrizian, Armin Mehrabian, Peng Wei 9d ago

EO-Agents: A Three-Agent LLM Pipeline for Earth Observation Hypothesis Generation

EO-Agents uses a three-agent LLM pipeline grounded in NASA Earth Observation Knowledge Graph with graph neural networks to generate scientific hypotheses from structured data.

Ax Junyi Wen, Ruiyan Zhuang, Yongjia Xu, Pengtu Li, Rui Zou, Hongyi Chen, Chingman Wan, Puxu Yang, Wuhui Chen, Yanlin Wang 9d ago

Hawk: Harnessing Hardware-Aware Knowledge for High-Performance NPU Kernel Generation

Hawk uses LLMs with hardware-aware knowledge and semantic understanding to generate high-performance kernels for Neural Processing Units, addressing implicit constraints.

Ax Junyan Tan, Haoran Lin, Siyuan Guo, Yichen Fang, Xinyue Luo, Tianyu Shen, Zeyu Qiao 9d ago

Safe and Adaptive Cloud Healing: Verifying LLM-Generated Recovery Plans with a Neural-Symbolic World Model

Neural-symbolic world model approach for cloud system fault detection and recovery using LLMs for semantic understanding and DRL for policy optimization.

Ax Xinyi Fang, Kejian Tong, Jiabei Liu, Tao Ning, Yuhang He 9d ago

SemHash-LLM: A Multi-Granularity Semantic Hashing Framework for Document Deduplication

SemHash-LLM framework combines semantic hashing, MinHash, contrastive learning, and LLM-based filtering for efficient large-scale document deduplication while preserving semantic equivalence.

Ax Keita Kinjo, Takeshi Ebina 9d ago

Profit-Based Counterfactual Explanations for Product Improvement: A Case Study of Manga Sales in Japan

Counterfactual explanation method for ML interpretability applied to manga sales prediction, addressing target specification and distance function challenges.

Ax Xuqing Yang, Yi Yuan, Shanzhe Lei, Xuhong Wang 9d ago

Scaling with Confidence: Calibrating Confidence of LLMs for Adaptive Test Time Scaling

Research on calibrating LLM confidence via RL to accurately express uncertainty during test-time scaling for reasoning and QA tasks.

Ax Low Jun Yu, Niramay Kachhadiya, Herath Mudiyanselage Viraj Vidura Herath, Sanka Rasnayaka, Lucy Amanda Marshall 9d ago

Spatial Support Matters: Geometry-Aware Graph Fusion for Rainfall Field Reconstruction

Graph fusion method for rainfall reconstruction reconciling point gauges, path microwave links, and gridded radar/satellite with geometry-aware approach.

Ax Xingyuan Dai, Yue Liu, Xiaoyan Gong, Qinghai Miao, Junyou Shang, Yutong Wang, Chao Guo, Yonglin Tian, Yizhang Chai, Chao Xiang, Yisheng Lv, Fei-Yue Wang 9d ago

Autonomous discovery of traffic laws with AI traffic scientists

Autonomous AI system discovers universal traffic laws from observational data to identify recurrent congestion patterns across cities.

Ax Yuante Li, Yicheng Tao, Kate Zhang, Taozhi Wang, Gefei Gu, Yaxin Zhou 9d ago

Diverse Evidence, Better Forecasts: Multi-Agent Deliberation Under Information Asymmetry

Multi-agent LLM deliberation improves forecasting when agents receive diverse evidence; identical evidence causes herding rather than genuine belief revision.

Ax Yufan Lu, Xinhui Liu, Chenyang Xu, Yuxi Zhou, Hao Wang, Shenda Hong 9d ago

Separating Expert Retention from Autonomous Source Inference in Raw-ECG-Replay-Free Continual ECG Deployment

Research on continual ECG model deployment separating expert retention from source inference using frozen backbone and isolated classifiers per source.

Ax Joshua Penman 9d ago

Epistemic Goggles: A Pretrained Module that Induces an Epistemic Frame via Gradient Editing

Goggles is a learned module using gradient editing to fix Negation Neglect, where finetuned LLMs fail to recognize fictional content (9% accuracy baseline).

Ax Zongxia Li, Dawei Liu, Fuxiao Liu, Yuhang Zhou, Xiyang Wu, Jingxi Chen, Jing Xie, Xiaomin Wu, Lichao Sun 9d ago

COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows

COMFYCLAW introduces self-evolving skill harnesses for agentic workflow-based image generation, enabling agents to recall patterns and preferences.

Ax Yongqin Zeng, Sicheng Pan, Jiale Wang, Hai-tao Zheng, Hong-Gee Kim, Chunxia Ma, XiuTeng Zhou 9d ago

Generic Expert Coverage for Pruning SparseMixture-of-Experts Language Models

Generic TB-Coverage proposes coverage-aware expert pruning for sparsely-activated Mixture-of-Experts language models without downstream calibration data.

Ax Xudong Wu, Jian Qian, Pangpang Liu, Vaneet Aggarwal, Jiayu Chen 9d ago

Distributionally Robust Listwise Preference Optimization

Distributionally robust listwise preference optimization for LLM alignment under ranking-label uncertainty from annotator inconsistency and noise.

Ax Yueming Huang, Wenhan Yao, Fen Xiao, Xiarun Chen, Weiping Wen 9d ago

DRL-CLBA: A Clean Label Backdoor Attack for Speech Classification via DDPG Reinforcement Learning

DRL-CLBA proposes clean-label backdoor attack on speech classification models using DDPG reinforcement learning.

Ax Simon Guilloud, Sankalp Gambhir, Samuel Chassot 9d ago

Reformalization of the Jordan Curve Theorem

Case study in reformalization converting Jordan Curve Theorem proofs between proof assistants: Mizar, HOL Light, Lean, and Agda.

Ax Blair Hudson 9d ago

Meta-Benchmarks for Financial-Services LLM Evaluation

Meta-benchmarking framework organizing 452 public benchmarks into 41 O*NET work activities for evaluating financial-services LLM performance.

Ax Sung June Kim, Sangpil Kim, Honglak Lee 9d ago

Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation

Phi-Nav addresses semantic drift in vision-language navigation by generating hindsight instructions for on-policy exploration training.

Ax Mingzhe Du, Luu Anh Tuan, Tianyi Wu, Renyang Liu, Zhijiang Guo, Dong Huang, See-Kiong Ng 9d ago

Mastermind: Strategy-grounded Learning for Repository-Scale Vulnerability Reproduction

Mastermind trains agents to reproduce vulnerabilities at repository scale by learning strategies for codebase inspection, input grammar inference, and PoC construction.

Ax Chunjiang Liu, Xiaoyuan Wang, Haoyu Chen, Yizhou Zhao, Ming-Hsuan Yang, L\'aszl\'o A. Jeni 9d ago

SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation

SimWorlds enables LLM agents to generate dynamic 4D scenes with physics simulation from text, supporting liquids, particles, and articulated motion.

Ax Xinyuan Song, Zekun Cai 9d ago

Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts

Addresses world-model correction in long-horizon agent planning by fixing underlying issues rather than replanning entire graphs after mistakes.

Ax Yujin Yang, Heejung Lee 9d ago

Verifiable Knowledge Expansion through Retrieval-Grounded Formal Concept Analysis

Proposes retrieval-augmented SLM framework using formal concept analysis as symbolic verification loop for ontology and knowledge structure expansion.

Ax Maximo Rulli (Sapienza University of Rome), Thomas Fontanari (Sapienza University of Rome), Simone Petruzzi (Sapienza University of Rome), Federico Alvetreti (Sapienza University of Rome), Giorgio Strano (Sapienza University of Rome), Donato Crisostomi (Sapienza University of Rome), Giorgos Nikolaou (EPFL), Tommaso Mencattini (EPFL), Andrea Santilli (Independent researcher), Emanuele Rodol\`a (Sapienza University of Rome), Simone Scardapane (Sapienza University of Rome), Alessio Devoto (Independent researcher) 9d ago

Subliminal Clocks: Latent Time Modelling in Diffusion Language Models

Studies latent time representations in Diffusion Language Models, showing they internally encode denoising progress information without explicit timestep conditioning.

Ax Yunhao Feng, Ruixiao Lin, Ming Wen, Qinqin He, Yanming Guo, Yifan Ding, Yutao Wu, Jialuo Chen, Yunhao Chen, Xiaohu Du, Jianan Ma, Zixing Chen, Zhuoer Xu, Xingjun Ma, Xinhao Deng 9d ago

Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification

Vera is an automated safety testing framework for LLM agents performing autonomous actions, scaling risk discovery beyond expert-designed violations.

Ax Lihui Luo, Joongwon Chae, Ziyan Chen, Yang Liu, Siyi Cheng, Weihan Gao, Zelin Zeng, Xiaoming Yin, Samaneh Beheshti Kashi, Dongmei Yu, Lian Zhang, Jing Sui, Zeming Liang, Jiansong Ji, Peter E. Lobie, Peiwu Qin 9d ago

MMIR-TCM: Memory-Integrated Multimodal Inference and Retrieval for TCM Clinical Decision Support

MMIR-TCM applies multimodal AI to Traditional Chinese Medicine diagnosis and clinical decision support using memory-integrated retrieval.

Ax Alex Brooker, Tim Hughes 9d ago

Pre-Flight: A Benchmark for Evaluating Large Language Models on Aviation Operational Knowledge

Pre-Flight is an open-source benchmark of 300 multiple-choice questions evaluating LLMs on aviation-specific operational knowledge and safety reasoning.

Ax Georgiana Caltais, Milan Lopuha\"a-Zwakenberg, Mari\"elle Stoelinga 9d ago

Actual causality in fault trees

Applies Halpern & Pearl's actual causality theory to fault trees for failure diagnostics in complex systems.

Ax Fangfei Li, Chenyang Zhao, Long Wang, Feng Tian, Zhiyue Zheng, Lv Guo 9d ago

CLAP: Closed-Loop Training, Evaluation, and Release Control for Domain Agent Post-training

CLAP presents closed-loop methodology for domain agent post-training, converting business data into structured samples with validation and release-gate controls.

Ax Joshua Adrian Cahyono 9d ago

Safety Targeted Embedding Exploit via Refinement

STEER identifies safety gaps in LLM multilingual and code-switching inputs, revealing models generate harmful responses outside English safety training distribution.

Ax Dawei Ren, Yan Zhang, Hongying Tang, Qiaoling Zhou, Jianpo Liu 9d ago

CamoNAS: Neural Architecture Search for Enhanced Camouflaged Object Detection

CamoNAS uses neural architecture search with frequency-aware multi-resolution framework to improve camouflaged object detection in images.

Ax Jiayin Zhu, Kelong Mao, Yudong Guo, Dengbo He, Sulong Xu, Simiu Gu, Yutao Yue 9d ago

SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use

SkillCoach proposes self-evolving rubrics for evaluating and training LLM agents to reliably use skills from repositories with overlapping capabilities.

Ax Tianjian Yang, Meng Li 9d ago

Spec-AUF: Accept-Until-Fail Training under Train-Inference Misalignment for Masked Block Drafters

Spec-AUF training method for masked block drafters in speculative decoding addressing train-inference misalignment.

Ax Zihao Xu, Yuekang Li, Gelei Deng, Yi Liu, Zhenchang Xing 9d ago

Rethinking Complexity Metrics for LLM-Integrated Applications: Beyond Source Code

HECATE tool for measuring complexity in LLM-integrated applications across both prompt and code layers.

Ax Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia, Wei Yang, Xiuchang Li, Jinwei Zhu, Yang Ren, Lei Chen, Gao Cong 9d ago

ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair

ContextSniper: token-efficient code memory layer for LLM agents performing repository-level program repair with reduced context overhead.