Health
1353 storiesPublic health and medicine worldwide — sourced from WHO News.
VocalAffectBench: Evaluating Vocal Emotion Recognition in AI Audio Models
arXiv:2608.28932v2 Announce Type: replace Abstract: Voice products increasingly need affective cues that are present in speech but absent from transcripts. We introduce VocalAffectBench, a public, test-only benchmark for…
Automated Researchers Can Mitigate Well-characterized Alignment Failures
arXiv:2608.28945v3 Announce Type: replace Abstract: Automating alignment research may accelerate progress toward aligned AI, but whether it does is hard to measure. Luckily, many alignment failures, such as deception, sy…
DReSG: Diffusion Residuals for Stylized Gaussian Splatting
arXiv:2608.29048v2 Announce Type: replace Abstract: Reference-guided stylization of scenes represented by 3D Gaussian Splatting (3DGS) is important for efficient and controllable 3D content creation. Existing VGG-feature…
Mechanizing Typed Regulatory Actions for Security Tokens: Semantics, Falsification, and Bounded EVM Evidence
arXiv:2608.29134v2 Announce Type: replace Abstract: Security-token standards expose privileged controls without identifying the legal effect executed or the evidence and reversal obligations it carries. We formalize in I…
Hyper-Fold: Exploring the Expressive Limit of Sequence-Geometry Learning for Proteins via Hypergraph Modeling
arXiv:2608.29207v2 Announce Type: replace Abstract: Protein structure modeling rests on a single computational primitive: the interaction between what a residue is (sequence content) and where it sits (three-dimensional …
Attribute-Based Activation Steering of LLMs for Group-Specific Explanation Generation
arXiv:2608.29215v2 Announce Type: replace Abstract: To effectively enable people to understand new topics, explanations should be tailored to their backgrounds and abilities. Prompting alone has been shown to be insuffic…
AnyWorld: Factorized Egocentric World Models for Cross-Embodiment Generalization
arXiv:2608.29242v2 Announce Type: replace Abstract: Collecting contact-rich robot experiences at scale remains a major bottleneck for generalizable manipulation. Beyond data quantity, robot learning also requires diverse…
Validating FKG.in: Soundness Assessment in LLM-Augmented Indian Food Knowledge
arXiv:2608.29249v2 Announce Type: replace Abstract: The online culinary ecosystem is increasingly populated by recipe content generated, modified, or summarized by Large Language Models (LLMs). While often plausible, suc…
Accelerating Unified Multimodal Models with Core-Expansion Routing and Unified Computation Scheduling
arXiv:2608.29291v3 Announce Type: replace Abstract: Unified multimodal models jointly support understanding and generation, but incur substantial redundant computation across tokens, layers, and generation timesteps. Thr…
Chain-of-Thought Faithfulness of Reasoning Models Varies with Where and How Preference Cues Are Delivered
arXiv:2608.29464v2 Announce Type: replace Abstract: Chain-of-thought (CoT) monitoring assumes that reasoning traces faithfully record the information that shapes a model's answer. Existing faithfulness tests often place …
TACS: Trajectory-Aware Candidate Selection for LLM Jailbreak Suffix Optimization
arXiv:2608.29564v2 Announce Type: replace Abstract: Gradient-based jailbreak suffix optimization methods typically update the suffix by retaining the candidate with the lowest current loss. We show that this seemingly na…
OASIS: Optimizing Attacker Sequences for Hard-Label Black-Box Text Attacks
arXiv:2608.29568v2 Announce Type: replace Abstract: Different attack methods follow different search trajectories, they succeed on different subsets of samples, whereas existing hard-label black-box text attacks mainly f…
HiVe: Beyond Static Prompts for Multitask Learning via Hierarchy-based Vertical Mixture-of-Experts
arXiv:2608.29790v2 Announce Type: replace Abstract: As large language models (LLMs) continue to scale, parameter-efficient fine-tuning (PEFT) has become a practical alternative to full-parameter adaptation. Prompt tuning…
When History Is Multimodal: Rethinking Context Management for Long-Horizon Agents
arXiv:2608.29897v2 Announce Type: replace Abstract: Long-horizon agents need a context manager to compress growing interaction histories into a bounded working context, via passive strategies or active strategies that de…
REIGN: Refurbished Embeddings with Integrated Guidance Networks for Efficient Context-Length Scaling
arXiv:2608.29899v2 Announce Type: replace Abstract: Dense retrieval over long documents is expensive. Token-level encoders scale quadratically in sequence length, and most long-context embedding models reach 32K tokens o…
XQDT: eXplainable and Quantitative Data-Text Alignment Metric with Feedback Signals
arXiv:2608.29948v2 Announce Type: replace Abstract: Evaluating data-text alignment remains challenging: existing metrics often provide limited explanations for the scores, while prompt-based LLM-as-Judge methods can be e…
Arkios: An Open Bilingual English-Nepali Language Model Trained From Scratch, with a Devanagari-Aware Tokenizer
arXiv:2608.30092v2 Announce Type: replace Abstract: We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single-file C/CUDA traini…
E-SENS: Exclusion-Sensitive Penalization for Negative-Constraint Retrieval
arXiv:2608.30130v2 Announce Type: replace Abstract: Retrieval-augmented language models can fail to respect negative constraints when the retriever supplies evidence about concepts the user explicitly excluded. Beyond ex…
DICS: Exploring Data Intrinsic Consistency for Visual Instruction Selection
arXiv:2608.30209v3 Announce Type: replace Abstract: Visual instruction tuning is crucial for advancing the vision-language alignment and instruction-following capabilities of Vision-Language Models (VLMs). However, ident…
Lazy Grounding: Attacking Search Agents with Factual Evidence
arXiv:2608.30303v2 Announce Type: replace Abstract: Search agents mitigate hallucination by grounding their answers in retrieved web results. However, retrieval-based approaches also introduce an attack surface: agents m…