New papers on fairness, safety, alignment and governance.

LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing age

Coding agents have made substantial progress on software engineering tasks that modify existing codebases, including bug fixing and feature implementation. However, constructing a complete program from scratch remains a major challenge: even the frontier models evaluated on ProgramBench fully resolve fewer than 1% of tasks. One obstacle is the lack of scalable training environments for this from-scratch setting, spanning the whole software engineering life cycle, as existing environment-construc

Existing autoregressive video distillation methods commonly adopt a Distribution Matching Distillation (DMD)-based multi-stage pipeline. However, they typically decouple the initialization and DMD stages -- which then pursue different target distributions -- and judge the intermediate student mainly by visual scores such as VBench. In this paper, we revisit this design from a distributional perspective. Given the mode-seeking nature of the distribution matching loss, a good initialization should

We present Voice Memory, a inference-only scheme for agentic speech recognition: at stream time, a frozen corrector reads a single per-domain memory.md and decides per utterance whether to act on the hypothesis or abstain and keep the 1-best. Asynchronously, a score-gated optimizer revises that file through bounded edits, accepting an edit only when it strictly improves a held-out score. Extended from classical ASR-LM framework, we refer this split the listener-thinker architecture; the two role

Recent game world models can generate visually realistic and interactive environments conditioned on player actions. However, games are not defined by pixels alone; they are governed by explicit mechanics, namely state-dependent rules that control health reduction, skill activation, and game termination. These mechanics depend on precise internal states, such as health points, skill meters, and timers, which are tightly coupled with visual observations and determine how gameplay evolves. Without

Large Language Model (LLM) agents are increasingly adopted in real-world security operations with access to host artifacts and command-line interfaces (CLIs), making it critical to thoroughly assess their security capabilities. However, existing cybersecurity benchmarks focus on pre-compromise settings where agents are placed in a clean and idealized environment before an attack occurs. This leaves the post-compromise setting underexplored. To address this gap, we introduce SecRespond, the first

Fact-checking saw a rapid expansion in the mid 2010s when major social media platforms, especially Meta, started funding these activities. Today, however, fact checkers stand at a critical juncture. The early 2020s brought them three interrelated yet distinct crises: financial, technological, and legitimacy crises. The post The future of fact-checking in the algorithmic society first appeared on HKS Misinformation Review .

In this study, we surveyed users for their perception of misinformation exposure across social media platforms. Such perceived exposure is important because individuals' beliefs about how often they encounter false information can shape their trust in institutions, platforms, and even their friends. In a survey of 1,010 United States residents, we found that perceived exposure to misinformation varies substantially across platforms and is only moderately correlated with the frequency of platform

Multi-Agent Debate (MAD) is a promising paradigm for improving the accuracy and robustness of Large Language Model (LLM)-based agentic systems. It enables multiple agents to exchange arguments, critique each other's outputs, and iteratively converge towards a solution. However, research remains fragmented, with inconsistent terminology and no rigorous synthesis of MAD design dimensions. We present a systematic literature review characterizing 141 primary studies on MAD. We derive a three-dimensi

Background: Historic and ongoing problematic practices regarding the collection, storage, and use of Indigenous health data have led to the need to ensure principles of Indigenous Data Sovereignty (IDS) are followed in research practices and technology development. Objective: This project, a partnership between UC San Diego and the Native BioData Consortium (NativeBio), sought to explore the practical application of blockchain technology and its potential to facilitate Indigenous-led research co

A decentralized autonomous organization (DAO) is a governance entity that allows its stakeholders to manage blockchain-based protocols through smart contracts. The DAO explicitly specifies how stakeholders make and enforce decisions concerning a protocol's operation in a smart contract, aptly referred to as its governance contract. The design of this governance contract, therefore, has far-reaching implications for the security (trust) and privacy (transparency) of the smart contracts managed by

Background: Multimorbidity involves heterogeneous disease combinations, treatment burden, competing priorities, and complex care pathways. Digital health interventions (DHIs) may support monitoring, self-management, and care coordination, but their effects on health-related outcomes remain uncertain. Objective: This systematic review and meta-analysis evaluated the effectiveness of DHIs on physiological, psychological, and functional outcomes in adults with multimorbidity, summarized implementat

Background: Although thyroid nodules are detected in up to 60% of adults on ultrasound, the vast majority are benign, creating a substantial decision-making burden compounded by heterogeneous practice guidelines. Large language models (LLMs) show promise in processing unstructured medical text and are emerging as tools for report interpretation among both clinicians and patients. However, their reliability across distinct clinical tasks in thyroid ultrasound interpretation remains poorly charact

Alignment training, model organisms, and toy models are usually treated as separate research areas. But projects in all three frequently use supervised fine-tuning (SFT) to pursue the same underlying goals. When projects share a goal, we should test whether lessons learned from one area transfer to the other areas. We study three such transfers, each taking a lesson developed in one SFT setting and testing it in another. First, we port a lesson about behavior generalization from alignment traini

Automated molecular structure elucidation from infrared (IR) spectroscopy data has seen significant advancements in recent years, but its broad applicability is limited by a reliance on pre-determined chemical formulas provided as auxiliary model inputs. This limits model predictions to isomer identification rather than full molecular structure prediction. Although transformer models have been shown to identify molecular isomers with high accuracy, their reliability for unconstrained structure e

An AI benchmark result rarely reaches a consequential claim in one step. Evaluators generalize it to further cases, interpret it as evidence of capability, extrapolate it to new tasks, transport it to another system or site, and combine it with assumptions about human review and downstream consequences. Validity-centred approaches require evidence for each claim. This paper identifies a further epistemic problem: warranted links don't automatically make a warranted chain. The target of one study

Background: With the expanding need for advance care planning (ACP), innovative educational strategies for training health care providers are increasingly required. Large language model (LLM)–based ACP chatbots offer a novel and potentially effective solution to enhance health care providers’ competence in navigating complex ACP conversations. Objective: This study aimed to develop a Chinese-context ACP corpus to support an LLM-based chatbot and evaluate the feasibility and performance of a mult

On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix failure: once the student commits to a wrong reasoning direction, all subsequent generation builds on this deviation, producing misdirected continuations that elicit unreliable supervision and waste compute. We identify a teacher-student continuation asymmetry on failed prefixes, where the teacher tends to redirect while the student continues along the original direction, and con

Generalist manipulation policies increasingly take the form of action-chunking flow policies built on large pretrained backbones. Such chunks run open-loop, so the policy cannot react to sensory input arriving mid-execution, sacrificing \emph{reactivity}. Replanning more often would restore it, but the perception-to-action pipeline (a large backbone plus multiple denoising steps) is too slow: this \emph{latency} forbids frequent replanning and leaves committed actions stale, making such policies

Embodied intelligence integrates learned perception and decision making with real-time computation, control, and physical interaction. Because failures can cause immediate physical or operational harm, task completion alone does not establish trustworthiness. We define trustworthy embodied intelligence as the sustained capacity to execute specified tasks reliably under environmental and system variation while maintaining risk within acceptable bounds. We term this objective sustained safe succes

Computer-use agents (CUAs) increasingly act through desktop GUIs to complete long-horizon tasks. Current benchmarks primarily measure end-task success or single-frame grounding. Neither isolates whether a model can reconstruct the causal, task-relevant transition produced by an action- crucial for rejecting stale observations, verifying progress, and recovering from failure. This is difficult because inference, remote input, app rendering, and screenshot capture are asynchronous: the next observ

Recently, photonic transformer accelerators (PTAs) have successfully achieved significant speedup and energy efficiency improvements over electronic accelerators for expediting Transformer inference. However, state-of-the-art rely on expensive multi-wavelength light generation and large dot-product units due to active phase-shifter components, thus making their approach inefficient and impractical. To address this, we propose MDTransformer, a novel hardware-software co-design of PTA based on mod

Self-play in simulation produces robust driving policies at scale. Demonstrations of such behavior have been made using privileged vectorized observations such as exact poses and velocities, even for occluded agents. This assumes that perception is solved and introduces a representation gap with the partial observation of a deployed agent driving from the perspective view of egocentric cameras. A common fix, distilling the privileged policy into a camera-input student, leaves the student imitati

Background: The rapid emergence of artificial intelligence (AI) has outpaced its formal adoption in health care organizations, contributing to the emergence of Shadow AI, defined here as the use of unauthorized AI tools by medical professionals. Under the European Union Medical Device Regulation, AI tools used for clinical purposes must undergo conformity assessment before use; general-purpose tools such as ChatGPT have not done so, rendering their clinical application unauthorized at the regula

Background: Therapist-guided internet-delivered cognitive behavioral therapy (guided iCBT) is increasingly implemented in routine mental health care to expand access to evidence-based treatments. Although the clinical effectiveness and patient acceptability of guided iCBT for common mental health disorders are well established, less is known about how introducing such digitally mediated interventions reshapes therapists’ everyday work practices and professional roles. Existing research has prima

Recently, memory management has become a key infrastructure for LLM-based agents, as it directly affects long-horizon reasoning, personalized responses, and knowledge reuse. However, existing LLM memory systems typically adopt a coarse-grained (utility-agnostic) manner that treats heterogeneous user-LLM interaction records uniformly, leading to redundant and low-impact records persisting in the memory repository. To address this challenge, we present MemLens, a value-aware memory management syst

Exact-equivariant architectures typically encode prescribed group actions in specialized operators, which can complicate their reuse with generic backbones and across data modalities. We introduce the Generator-Aligned Representation Interface (GARI), a representation-level design principle that exposes selected transformation generators to a generic sequence backbone through aligned canonical and generator-induced views. We formalize the resulting behavior using a probe-specific soft-equivarian

Scientist have developed a host of AI ‘protein language models’ that can create proteins from scratch. But Raygun can modify existing proteins, using some of the same steps as natural evolution: ...

Scientific research involving code has long rested on the assumption that at least one person understands why the code exists. As scientists adopt coding agents, this assumption is breaking down. Drawing on an ongoing contextual inquiry of scientific programmers working with agentic tools (four cases to date), a survey of over 800 scientific programmers, and my own analysis workflows, this position piece describes how scientists are inventing personal conventions, "landmarking strategies", for m

Observed standardized test scores are the result of an endogenous process: students strategically allocate effort across multiple retake attempts to improve their outcomes. Because students differ in their ability to make these investments, the interaction between applicant strategy and institutional scoring rules---such as the widely used Single-Sitting and Superscoring policies---can disparately distort observed scores. We develop a strategic framework where students allocate effort in respons

EEG-based community sensing programs are emerging globally as a tool to leverage aggregated brain data to gain insights into attentiveness of students and employees. But these programs raise privacy concerns because EEG signals contain sensitive personal information. Differential Privacy (DP) can protect individuals while preserving aggregate statistics yet applying DP to EEG data is challenging as it requires user-level noise generation, which increases power and latency. Besides, most commerci

Ambivalence and hesitancy (A/H) are conflicting affective states that precede the delay or abandonment of health behaviour change. Recognition of A/H at the video level is difficult, since the signal arises from disagreement across and within facial, vocal, linguistic, and bodily modalities, and manifests differently across individuals. The proposed PRISM-AH (Predictive Reasoning over Interacting Streams for Multimodal Ambivalence/Hesitancy Recognition), is a framework that treats A/H as a multi

Ambivalence and hesitancy (A/H) are conflicting affective states that precede the delay or abandonment of health behaviour change. Recognition of A/H at the video level is difficult, since the signal arises from disagreement across and within facial, vocal, linguistic, and bodily modalities, and manifests differently across individuals. The proposed PRISM-AH (Predictive Reasoning over Interacting Streams for Multimodal Ambivalence/Hesitancy Recognition), is a framework that treats A/H as a multi

Any-to-any models predict any modality from any combination of others within a single network, a formulation used in multimodal vision and vision-language models, and increasingly in scientific domains such as ecology and astronomy. Existing any-to-any models are typically trained from scratch using encoder-decoder or diffusion architectures, impacting their performance and preventing them from using strong pre-trained decoder-only models as a prior. In this work, we investigate decoder-only any

Question answering (QA) over irregular clinical time series (ICTS) plays a pivotal role in a wide range of healthcare applications. Although recent multimodal time-series large language models (LLMs) have shown considerable promise in general-purpose time-series QA, they remain poorly equipped to model the sparsity, asynchrony, and irregular sampling patterns of clinical observations. To fill this gap, we propose ClinPRISM, a cost-effective multimodal LLM reasoning framework for question answeri

Clinical diagnostic evaluation should not only assess whether models can provide correct diagnoses, but also reflect the realities of clinical practice, including progressive disclosure of multimodal information, dynamic updating of diagnostic hypotheses, and continuous refinement of clinical reasoning. However, existing evaluations of multimodal large language models (MLLMs) typically rely on single-turn or isolated tasks, making it difficult to fully capture the complexity of real-world clinic

Face de-identification (De-ID) aims to remove or conceal personally identifiable facial features in images or videos to prevent identity recognition while preserving utility for downstream tasks. With the rising emphasis on data privacy and responsible AI, face De-ID has emerged as an active research area spanning computer vision and privacy-preserving communities. Early approaches, and many contemporary ones, operate in the digital domain by modifying pixel-level or appearance-level features th

Over the past decade, decision trees have been used to represent controllers (a.k.a. policies) in an explainable way, with dtControl2 as a current state-of-the-art tool. However, for systems that are large or have many corner cases, even such representations tend to be too complex and not human-comprehensible. Unfortunately, reducing the size of the decision tree is not straightforward, as missing just a single crucial case might result in an incorrect controller. We tackle this issue in the set

The rise of generative artificial intelligence (GenAI) has fueled high expectations regarding its potential to enhance knowledge work productivity in terms of efficiency and quality. Building on task-technology fit (TTF) theory, we empirically examine the extent of GenAI's productivity effect for different task types. We conducted a randomized lab-in-the-field experiment with 128 knowledge workers from a multinational industrial organization. Participants completed three representative knowledge