Safety-aligned large language models (LLMs) are becoming increasingly widespread, especially in sensitive applications where fairness is essential and biased outputs can cause significant harm. However, evaluating the fairness of models is a complex challenge, and approaches that do so typically utilize standard question-answer (QA) styled schemes. Such methods often overlook deeper issues by interpreting the model's refusal responses as positive fairness measurements, which creates a false sense of fairness. In this work, we introduce the concept of silenced biases, which are unfair preferences encoded within models' latent space and are effectively concealed by safety-alignment. Previous approaches that considered similar indirect biases often relied on prompt manipulation or handcrafted implicit queries, which present limited scalability and risk contaminating the evaluation process with additional biases. We propose the Silenced Bias Benchmark (SBB), which aims to uncover these biases by employing activation steering to reduce model refusals during QA. SBB supports easy expansion to new demographic groups and subjects, presenting a fairness evaluation framework that encourages the future development of fair models and tools beyond the masking effects of alignment training. We demonstrate our approach over multiple LLMs, where our findings expose an alarming distinction between models' direct responses and their underlying fairness issues.

The overall architecture of CFDT is shown in Figure 1. The base backbone is consistent with the network illustrated in the section of 3.1 Local-Global Cross Fusion. As shown by the red dotted lines in Figure 1, we use 100 det tokens as the additional input to perform self attention in the backbone. The det tokens dimension is also set as 256. Neck is a decoder-only modules, and there are 6 decoder layers in this neck.

When you purchase through links in our articles, we may earn a small commission. Google's new AI service turns into your own private tutor With Guided Learning, Google Gemini turns into your own educational tutor. When ChatGPT launched three years ago, it shook the academic world to its core. Suddenly, students could have AI answer questions and even write essays. And because ChatGPT is so articulate, spotting cheaters became increasingly difficult.

As OpenAI expands in every direction, the new CEO of Applications is on a mission to make ChatGPT indispensable and lucrative. In case OpenAI's structure couldn't get any weirder--a nonprofit in charge of a for-profit that's become a public benefit corporation--it now has two CEOs. There's Sam Altman, chief executive of the whole company, who manages research and compute. And as of this summer, there's Fidji Simo, the former CEO of Instacart, who manages everything else. Simo hasn't been seen much at OpenAI's San Francisco office since she began as CEO of Applications in August. But her presence is felt at every level of the company--not least because she's heading up ChatGPT and basically every function that might make OpenAI money. Simo is dealing with a relapse of postural orthostatic tachycardia syndrome (POTS) that makes her prone to fainting if she stands for long periods of time. "Being present from 8 am to midnight every day, responding within five minutes, people feel like I'm there and that they can reach me immediately, that I jump on the phone within five minutes," she tells me. Employees confirm that this is true. OpenAI's famously Slack-driven culture can be overwhelming for new hires. Employees say she is often seen popping into channels and threads, sharing thoughts and asking questions.

Optimizing the performance of large language models (LLMs) on large-scale AI training and inference systems requires a scalable and expressive mechanism to model distributed workload execution. Such modeling is essential for pre-deployment system-level optimizations (e.g., parallelization strategies) and design-space explorations. While recent efforts have proposed collecting execution traces from real systems, access to large-scale infrastructure remains limited to major cloud providers. Moreover, traces obtained from existing platforms cannot be easily adapted to study future larger-scale system configurations. We introduce Symbolic Tensor grAph GEnerator(STAGE), a framework that synthesizes high-fidelity execution traces to accurately model LLM workloads. STAGE supports a comprehensive set of parallelization strategies, allowing users to systematically explore a wide spectrum of LLM architectures and system configurations. STAGE demonstrates its scalability by synthesizing high-fidelity LLM traces spanning over 32K GPUs, while preserving tensor-level accuracy in compute, memory, and communication. STAGE is publicly available to facilitate further research in distributed machine learning systems: https://github.com/astra-sim/symbolic tensor graph

Newsrooms and journalists across the world are adopting Generative AI (GenAI). Drawing on in-depth interviews with 23 journalists, this study identifies Bangladeshi journalists' awareness, acceptance, usage patterns, and their media organizations' stance toward GenAI. This study finds Bangladeshi journalists' high reliance on GenAI like their Western colleagues despite limited institutional support and the near absence of AI policy. Despite this contrast, concerns over GenAI's implications in journalism between the West and non-West were mostly identical. Moreover, this study contributes to the Unified Theory of Acceptance and Use of Technology (UTAUT) by proposing two changes regarding GenAI adoption among journalists in non-Western settings. First, this study identifies the non-contribution of facilitating conditions in shaping behavioral intent in GenAI adoption in non-Western contexts. Second, social influence works in a horizontal order through informal peer pressure or professional motivation in the absence of formal institutional hierarchical pressure. Voluntariness in the context of Bangladeshi journalists is underpinned by their professional compulsion. Therefore, this study contributes to understanding how contextual factors shape technology adoption trajectories in non-Western journalism.

Unit testing in High-Performance Computing (HPC) is critical but challenged by parallelism, complex algorithms, and diverse hardware. Traditional methods often fail to address non-deterministic behavior and synchronization issues in HPC applications. This paper introduces HPCAgentTester, a novel multi-agent Large Language Model (LLM) framework designed to automate and enhance unit test generation for HPC software utilizing OpenMP and MPI. HPCAgentTester employs a unique collaborative workflow where specialized LLM agents (Recipe Agent and Test Agent) iteratively generate and refine test cases through a critique loop. This architecture enables the generation of context-aware unit tests that specifically target parallel execution constructs, complex communication patterns, and hierarchical parallelism. We demonstrate HPCAgentTester's ability to produce compilable and functionally correct tests for OpenMP and MPI primitives, effectively identifying subtle bugs that are often missed by conventional techniques. Our evaluation shows that HPCAgentTester significantly improves test compilation rates and correctness compared to standalone LLMs, offering a more robust and scalable solution for ensuring the reliability of parallel software systems.

Personal attacks have become a notable feature of U.S. presidential debates and play an important role in shaping public perception during elections. Detecting such attacks can improve transparency in political discourse and provide insights for journalists, analysts and the public. Advances in deep learning and transformer-based models, particularly BERT and large language models (LLMs) have created new opportunities for automated detection of harmful language. Motivated by these developments, we present a framework for analysing personal attacks in U.S. presidential debates. Our work involves manual annotation of debate transcripts across the 2016, 2020 and 2024 election cycles, followed by statistical and language-model based analysis. We investigate the potential of fine-tuned transformer models alongside general-purpose LLMs to detect personal attacks in formal political speech. This study demonstrates how task-specific adaptation of modern language models can contribute to a deeper understanding of political communication.

Instruction-guided text-to-speech (TTS) research has reached a maturity level where excellent speech generation quality is possible on demand, yet two coupled biases persist: accent bias, where models default to dominant phonetic patterns, and linguistic bias, where dialect-specific lexical and cultural cues are ignored. These biases are interdependent, as authentic accent generation requires both accent fidelity and localized text. We present Contextual Linguistic Adaptation and Retrieval for Inclusive TTS sYnthesis (CLARITY), a backbone-agnostic framework that addresses these biases through dual-signal optimization: (i) contextual linguistic adaptation that localizes input text to the target dialect, and (ii) retrieval-augmented accent prompting (RAAP) that supplies accent-consistent speech prompts. Across twelve English accents, CLARITY improves accent accuracy and fairness while maintaining strong perceptual quality.

Frontier model progress is often measured by academic benchmarks, which offer a limited view of performance in real-world professional contexts. Existing evaluations often fail to assess open-ended, economically consequential tasks in high-stakes domains like Legal and Finance, where practical returns are paramount. To address this, we introduce Professional Reasoning Bench (PRBench), a realistic, open-ended, and difficult benchmark of real-world problems in Finance and Law. We open-source its 1,100 expert-authored tasks and 19,356 expert-curated criteria, making it, to our knowledge, the largest public, rubric-based benchmark for both legal and finance domains. We recruit 182 qualified professionals, holding JDs, CFAs, or 6+ years of experience, who contributed tasks inspired by their actual workflows. This process yields significant diversity, with tasks spanning 114 countries and 47 US jurisdictions. Our expert-curated rubrics are validated through a rigorous quality pipeline, including independent expert validation. Subsequent evaluation of 20 leading models reveals substantial room for improvement, with top scores of only 0.39 (Finance) and 0.37 (Legal) on our Hard subsets. We further catalog associated economic impacts of the prompts and analyze performance using human-annotated rubric categories. Our analysis shows that models with similar overall scores can diverge significantly on specific capabilities. Common failure modes include inaccurate judgments, a lack of process transparency and incomplete reasoning, highlighting critical gaps in their reliability for professional adoption.