Source coverage · refreshed from the daily record

HuggingFace Daily Papers in the AI ethics record

A source-linked view of 410 research records gathered from HuggingFace Daily Papers. This page tracks what entered the ethics.ai source fleet; it is not a complete archive of the publisher and does not imply its endorsement.

Records by publication daylatest 90 days
2026-05-18 2026-08-15
410records in archive
410latest 90 days
43distinct publication days
12 August 2026latest published record

Most common automatic topics

Agents & autonomy 249
Regulation 82
Environment 68
Safety & alignment 38
Healthcare 33
Children & education 30
Finance, VC & PE 14
Transparency 13
Bias & fairness 12
Privacy 8

Source status and scope

last source check succeeded. The source is configured on a daily cadence and was last checked 50m ago.

Topic labels are automatic and can be imperfect. Counts measure records captured by ethics.ai, not everything the publisher produced, readership, importance or agreement with a claim.

Latest records from HuggingFace Daily Papers

All tracked sources →
HuggingFace Daily Papers

H2R-Bench: Benchmarking Human-to-Robot Manipulation Video Generation in World Models — open the original publisher

Large-scale manipulation data is essential for robot learning, yet collecting robot demonstrations remains expensive and difficult to scale. Meanwhile, abundant egocentric human manipulation videos provide rich behavioral experiences, but transferring them across embodiments remains challenging due to differences between human hands and robotic end-effectors. Recent advances in video world models offer a promising pathway to synthesize robot-centric manipulation videos from human observations, w

Research Agents & autonomy
HuggingFace Daily Papers

DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation — open the original publisher

We present DreamX-Phi 1.0, an action-conditioned video world model for robotic manipulation that, given an observed frame, a language instruction, and a prescribed action sequence comprising end-effector poses and gripper states, predicts the resulting future observations. Yet realism alone does not guarantee faithfulness: a convincing rollout can still move the wrong arm or lose the manipulated object. To ensure the prediction respects each arm's commanded path, we inject per-arm SE(3) transfor

Research Agents & autonomy
HuggingFace Daily Papers

Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence — open the original publisher

Spatial intelligence is becoming a foundation for embodied agents, robotic planning, and multimodal assistants. To improve the spatial reasoning ability of VLM agents, existing work has mainly followed two lines. One line uses post-training methods, such as supervised fine-tuning and reinforcement learning. Another line adopts an agentic paradigm in which the model calls external spatial tools, such as depth estimation and 3D reconstruction tools, to gather intermediate spatial evidence. We stud

Research Agents & autonomy
HuggingFace Daily Papers

AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design — open the original publisher

Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human design priors and accumulate reusable experience through empirical exploration to drive recursive self-improvement, existing paradigms remain static and fall short of this capability. In this paper, we present AutoDesign, a framework that aligns with human design

Research Agents & autonomy
HuggingFace Daily Papers

LycheeMemory V2: Efficient Long-Term Memory for LLM Agents via Semantic Segment-Level Consolidation — open the original publisher

Long-horizon LLM agents must preserve information from past interactions to support future tasks. Existing memory systems typically rely on eager consolidation, invoking LLMs after each interaction to extract, summarize, or update memories. This design makes memory construction increasingly costly as conversations grow. Coarse summarization can reduce construction cost but risks discarding fine-grained contextual evidence, whereas larger retrieval contexts or multi-hop LLM reasoning shift the ov

Research Agents & autonomy
HuggingFace Daily Papers

PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives — open the original publisher

Video world models simulate future states conditioned on current observations and user actions. Recent systems have demonstrated impressive video consistency and action controllability over long sequences. However, fairly comparing these interactive models remains challenging. In practice, a human player typically evaluates a world model by pursuing long-horizon objectives through interaction. For example, a user may turn around 360 degrees to see whether the environment remains consistent, or w

Research Agents & autonomyEnvironment
HuggingFace Daily Papers

Intern-S2-Preview: Scientific Agentic Foundation Model — open the original publisher

Scientific discovery increasingly requires AI systems that can reason over scientific evidence of heterogeneous modalities, interact with scientific tools and environments, and sustain progress across long task horizons. We present Intern-S2-Preview, a series of scientific agentic foundation models designed to support multimodal scientific understanding, reasoning, generation, and long-horizon tasks. The training pipeline begins with scientific multimodal pre-training over rendered scientific do

Research Agents & autonomyEnvironment
HuggingFace Daily Papers

Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation — open the original publisher

Interactive autoregressive video generation demands both low-latency rollouts and precise online control. Few-step distillation accelerates generation by reducing denoising steps, while online control imposes a causal constraint: frames and blocks should depend on history and controls available during generation. Existing video distribution matching distillation (DMD) pipelines, however, often supervise causal few-step students using bidirectional teachers that score complete clips. The score fo

Research Children & education
HuggingFace Daily Papers

Persistent Recursive Worlds Enable Autonomous Software Evolution — open the original publisher

Complex software systems develop over timescales that exceed the lifespan of any individual coding agent. Most agentic software systems preserve continuity through persistent sessions, memories, managers or shared context. We introduce EvoX Genesis (hereafter, Genesis), which instead makes the software project persistent while allowing local agents to remain finite-lived. Genesis represents software as a persistent recursive world: each local world is situated by an accepted version and a reposi

Research Agents & autonomy
HuggingFace Daily Papers

AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses — open the original publisher

Recent work on distillation transfers the capabilities of large models to smaller ones often by updating the latter's parameters, through teacher forcing, on-policy distillation, and related training-time methods. In this paper, we ask whether such transfer can instead occur at test time. We study strong-to-weak scaffolding: whether a stronger builder model can construct inference-time harnesses that help a weaker target model solve tasks more reliably without any parameter updates. Using four r

Research Regulation
HuggingFace Daily Papers

MBA: Multimodal Benchmark and Agents for Real-World Business Ideation — open the original publisher

Agentic systems powered by large language models (LLMs) have opened new opportunities for business ideation. Yet existing approaches remain confined to a text-only paradigm, despite the inherently multimodal nature of real-world contexts. We thus introduce MBA-Bench, the first multimodal benchmark for training and evaluating business ideation agents, comprising 30K samples across six domains, each domain characterized by distinct visual cues not fully conveyed by text alone. Concretely, we autom

Research Agents & autonomy
HuggingFace Daily Papers

AVA-Encoder: Towards Agent-Native Video Representation Learning — open the original publisher

Creative agents still lack an effective way to learn from high-quality human films, limiting their ability to produce cinematic-grade videos. A key challenge is the absence of a structured video representation that is both faithful to film content and directly usable for agentic reasoning and manipulation. To address the challenge, we propose the Agentic Video Auto-Encoder (AVA-Encoder), a framework for learning agent-native video representations via agentic auto-encoding. AVA-Encoder transforms

Research Agents & autonomy
HuggingFace Daily Papers

Spark-to-Paper: End-to-End Research Paper Generation as a Composable Skill — open the original publisher

Turning a research idea into a complete paper requires more than text generation: the system must retrieve literature, design and execute experiments, revise claims according to evidence, produce publication-ready figures, and maintain consistency across a long generation process. We present Spark-to-Paper, an end-to-end research paper generation system implemented as thirteen composable skills inside an existing coding assistant, without requiring a separate agent platform or orchestration serv

Research Agents & autonomy
HuggingFace Daily Papers

ToolHazard: Scaling Adversarial Environments for Security Evaluation and Alignment of LLM-based Agents — open the original publisher

Large language model (LLM) agents integrated with external tools are vulnerable to indirect prompt injections embedded in environmental states. However, existing studies largely rely on manually implemented or reused environments, stochastic LLM-based tool simulation, and predefined injection locations, limiting scalable security research across broader domains. To bridge this gap, we propose **ToolHazard**, a scalable adversarial environment synthesis framework that reduces human engineering an

Research Safety & alignmentAgents & autonomy
HuggingFace Daily Papers

Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence — open the original publisher

AI models have achieved remarkable success across diverse domains, yet the mechanisms underlying their capabilities and the risks they may pose remain poorly understood. As AI development becomes faster and increasingly automated, mechanistic exploration remains largely manual, widening the gap between what models can do and our ability to understand and control them. To bridge this gap, we introduce Mechanist, an agentic system that uses AI as a scientific instrument for the autonomous discover

Research Agents & autonomy
HuggingFace Daily Papers

Ready Cohorts: Bounding GPU Opportunity and Avoiding Host Round Trips in LLM-Agent Control — open the original publisher

LLM-agent services repeatedly execute small deterministic transitions between model and tool calls: route an outcome, update state, and emit the next effect. We ask when this control path exposes enough concurrent work for GPU execution, and what changes when a GPU-computed route decision remains on device. We formalize the ready-cohort boundary using fixed-partition share F, exact offline share P*, local upper bound U, and online achieved share A. Under zero service time, unlimited capacity, an

Research Agents & autonomy
HuggingFace Daily Papers

Hand Visibility Detector: Per-Keypoint Visibility Estimation for Hands — open the original publisher

Hand Pose Estimation (HPE) is a fundamental technology for various applications such as AR/VR and robotics. In these applications, the visibility of each hand joint in the image is crucial for assessing the reliability of estimation results under occlusion. However, most existing HPE methods output joint positions without explicitly indicating their visibility. Although some methods account for occlusion or visibility, visibility estimation has mainly been used as an auxiliary signal for improvi

Research Agents & autonomy
HuggingFace Daily Papers

Hybrid-Policy Self-Editing for Composable Unstructured Knowledge Editing — open the original publisher

Large language models (LLMs) achieve remarkable performance across natural language tasks, yet they are trained on static corpora and their knowledge quickly becomes outdated in a fast-changing world. This motivates knowledge editing (KE), which updates specific knowledge in an LLM without changing unrelated others. Recent works move from structured knowledge triples toward unstructured KE (UKE), where the edit is a free-form passage that may state multiple facts at once. Nonetheless, existing e

Research Regulation
HuggingFace Daily Papers

Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review — open the original publisher

This paper reports a single, fully instrumented case study of a large-scale architectural refactoring by an AI coding agent under a specification-first protocol, with no human review of the generated code and no pre-existing oracle to validate the target behaviour. The task, dismantling a central invariant across a large interdependent codebase, was assessed by the author as effectively infeasible through incremental refactoring, the kind of change that conventionally calls for a rewrite instead

Research Agents & autonomy
HuggingFace Daily Papers

Reference-Free Post-Training of Open Large Language Models for Multilingual Machine Translation — open the original publisher

We study reference-free post-training for multilingual machine translation with open large language models. Starting from the supervised-finetuned MiLMMT-46-v0.1 models, we apply Group Relative Policy Optimization (GRPO) with a reward that averages two reference-free quality estimation models and is gated by language identification. We then linearly interpolate the supervised fine-tuning (SFT) and reinforcement learning (RL) model checkpoints to obtain MiLMMT-46-v1.0. Across 46 languages, the re

Research Regulation

Method and reuse

ethics.ai stores source metadata, short summaries and links to the original publisher. It does not republish full articles. Use the permanent evidence link for citation, retain the original source link, and verify consequential claims with the publisher. See the methodology and corrections policy and reuse terms.