Static feature exclusion strategies often fail to prevent bias when hidden dependencies influence the model predictions. To address this issue, we explore a reinforcement learning (RL) framework that integrates bias mitigation and automated feature selection within a single learning process. Unlike traditional heuristic-driven filter or wrapper approaches, our RL agent adaptively selects features using a reward signal that explicitly integrates predictive performance with fairness considerations. This dynamic formulation allows the model to balance generalization, accuracy, and equity throughout the training process, rather than rely exclusively on pre-processing adjustments or post hoc correction mechanisms. In this paper, we describe the construction of a multi-component reward function, the specification of the agents action space over feature subsets, and the integration of this system with ensemble learning. We aim to provide a flexible and generalizable way to select features in environments where predictors are correlated and biases can inadvertently re-emerge.

Where an AI system is used, patients could face difficulties showing fault in the event of a negative outcome, experts say. Where an AI system is used, patients could face difficulties showing fault in the event of a negative outcome, experts say. The use of artificial intelligence in healthcare could create a legally complex blame game when it comes to establishing liability for medical failings, experts have warned. The development of AI for clinical use has boomed, with researchers creating a host of tools, from algorithms to help interpret scans to systems that can aid with diagnoses . AI is also being developed to help manage hospitals, from optimising bed capacity to tackling supply chains.

Safety alignment of large language models (LLMs) faces a key challenge: current alignment techniques often only focus on improving safety against harmful prompts, causing LLMs to become over-cautious and refuse to respond to benign prompts. Therefore, a key objective of safe alignment is to enhance safety while simultaneously reducing false refusals. In this paper, we introduce Energy-Driven Steering (EDS), a novel, fine-tuning free framework designed to resolve this challenge through dynamic, inference-time intervention. We trained a lightweight, external Energy-Based Model (EBM) to assign high energy to undesirable (false refusal or jailbreak) states and low energy to desirable (helpful response or safe reject) ones. We use the gradient of the energy function to dynamically steer the LLM's hidden states to low energy regions, correcting the model to generate a desirable response in real-time without modifying its weights. Extensive experiments across a wide range of models show our method successfully achieves this objective: it substantially lowers false refusal rates. Our work presents an effective paradigm for building LLMs that achieve both low false refusal rates and high safety. Note: This paper contains examples with potentially disturbing content. The alignment of large language models (LLMs) with human safety remains a central challenge in artificial intelligence research (Bianchi et al., 2023; Anwar et al., 2024; Xu et al., 2020; Röttger et al., 2020; Sun et al., 2021; Vidgen et al., 2023). Common approaches such as Supervised Fine-Tuning (SFT), Reinforcement Learning from Human Feedback (RLHF), system prompt engineering, and vector ablation have proven effective. However, these methods often introduce an unintended tradeoff: they can lead either to excessive refusal (over-rejection) or to lapses in safety. This behavior is not merely an inconvenience; it severely undermines model utility and reliability in critical domains.

Developing a high-performing large language models (LLMs) for low-resource languages such as Urdu, present several challenges. These challenges include the scarcity of high-quality datasets, multilingual inconsistencies, and safety concerns. Existing multilingual LLMs often address these issues by translating large volumes of available data. However, such translations often lack quality and cultural nuance while also incurring significant costs for data curation and training. To address these issues, we propose Alif-1.0-8B-Instruct, a multilingual Urdu-English model, that tackles these challenges with a unique approach. We train the model on a high-quality, multilingual synthetic dataset (Urdu-Instruct), developed using a modified self-instruct technique. By using unique prompts and seed values for each task along with a global task pool, this dataset incorporates Urdu-native chain-of-thought based reasoning, bilingual translation, cultural relevance, and ethical safety alignments. This technique significantly enhances the comprehension of Alif-1.0-8B-Instruct model for Urdu-specific tasks. As a result, Alif-1.0-8B-Instruct, built upon the pretrained Llama-3.1-8B, demonstrates superior performance compared to Llama-3.1-8B-Instruct for Urdu specific-tasks. It also outperformed leading multilingual LLMs, including Mistral-7B-Instruct-v0.3, Qwen-2.5-7B-Instruct, and Cohere-Aya-Expanse-8B, all within a training budget of under $100. Our results demonstrate that high-performance and low-resource language LLMs can be developed efficiently and culturally aligned using our modified self-instruct approach. All datasets, models, and code are publicly available at: https://github.com/traversaal-ai/alif-urdu-llm.

Refusal on harmful prompts is a key safety behaviour in instruction-tuned large language models (LLMs), yet the internal causes of this behaviour remain poorly understood. We study two public instruction tuned models--Gemma-2-2B-IT and LLaMA-3.1-8B-IT using sparse autoencoders (SAEs) trained on residual-stream activations. Given a harmful prompt, we search the SAE latent space for feature sets whose ablation flips the model from refusal to compliance, demonstrating causal influence and creating a jailbreak. Our search proceeds in three stages: 1. Refusal Direction - Finding a refusal mediating direction and collecting SAE features close to that direction, followed by 2. Greedy Filtering - to prune this set to obtain a minimal set and finally 3. Interaction Discovery - a factorization-machine (FM) model that captures non-linear interactions among the remaining active features and the minimal set. This pipeline yields a broad set of jailbreak-critical features, offering insight into the mechanistic basis of refusal. Moreover, we also find evidence of redundant features which remain dormant unless earlier features are suppressed. Our findings highlight the potential for fine-grained auditing and targeted intervention in safety behaviours by manipulating the interpretable latent space.

Large language model (LLM) agents often struggle in environments where rules and required domain knowledge frequently change, such as regulatory compliance and user risk screening. Current approaches, like offline fine-tuning and standard prompting, are insufficient because they cannot effectively adapt to new knowledge during actual operation. To address this limitation, we propose the Adaptive Reflective Interactive Agent (ARIA), an LLM agent framework designed specifically to continuously learn updated domain knowledge at test time. ARIA assesses its own uncertainty through structured self-dialogue, proactively identifying knowledge gaps and requesting targeted explanations or corrections from human experts. It then systematically updates an internal, timestamped knowledge repository with provided human guidance, detecting and resolving conflicting or outdated knowledge through comparisons and clarification queries. We evaluate ARIA on the realistic customer due diligence name screening task on TikTok Pay, alongside publicly available dynamic knowledge tasks. Results demonstrate significant improvements in adaptability and accuracy compared to baselines using standard offline fine-tuning and existing self-improving agents. ARIA is deployed within TikTok Pay serving over 150 million monthly active users, confirming its practicality and effectiveness for operational use in rapidly evolving environments.

Large Language Models (LLMs) and other automated techniques have been increasingly used to support software developers by generating software artifacts such as code snippets, patches, and comments. However, accurately assessing the correctness of these generated artifacts remains a significant challenge. On one hand, human evaluation provides high accuracy but is labor-intensive and lacks scalability. On the other hand, many automatic evaluation metrics are scalable and require minimal human effort, but they often fail to accurately reflect the actual correctness of generated software artifacts. In this paper, we present SE-Jury, the first evaluation metric for LLM-as-Ensemble-Judge specifically designed to accurately assess the correctness of generated software artifacts. SE-Jury first defines five distinct evaluation strategies, each implemented by an independent judge. A dynamic team selection mechanism then identifies the most appropriate subset of judges as a team to produce a final correctness score through ensembling. We evaluate SE-Jury across a diverse set of software engineering (SE) benchmarks that span three popular SE tasks: code generation, automated program repair, and code summarization. Results demonstrate that SE-Jury consistently achieves a higher correlation with human judgments, with improvements ranging from 29.6% to 140.8% over existing automatic metrics. SE-Jury reaches agreement levels with human annotators that are close to inter-annotator agreement in code generation and program repair. These findings underscore SE-Jury's potential as a scalable and reliable alternative to human evaluation in these SE tasks.

The performance of Large Language Models (LLMs) is determined by their training data. Despite the proliferation of open-weight LLMs, access to LLM training data has remained limited. Even for fully open LLMs, the scale of the data makes it all but inscrutable to the general scientific community, despite potentially containing critical data scraped from the internet. In this paper, we present the full-text indexing pipeline for the Apertus LLM training data. Leveraging Elasticsearch parallel indices and the Alps infrastructure, a state-of-the-art, highly energy-efficient arm64 supercluster, we were able to index 8.6T tokens out of 15.2T used to train the Apertus LLM family, creating both a critical LLM safety tool and effectively an offline, curated, open web search engine. Our contribution is threefold. First, we demonstrate that Elasticsearch can be successfully ported onto next-generation arm64-based infrastructure. Second, we demonstrate that full-text indexing at the scale of modern LLM training datasets and the entire open web is feasible and accessible. Finally, we demonstrate that such indices can be used to ensure previously inaccessible jailbreak-agnostic LLM safety. We hope that our findings will be useful to other teams attempting large-scale data indexing and facilitate the general transition towards greener computation.

Training expert LLMs in domains with scarce data is difficult, often relying on multiple-choice questions (MCQs). However, standard outcome-based reinforcement learning (RL) on MCQs is risky. While it may improve accuracy, we observe it often degrades reasoning quality such as logical consistency. Existing solutions to supervise reasoning, such as large-scale Process Reward Models (PRMs), are prohibitively expensive. To address this, we propose CLARity, a cost-effective RL framework that enhances reasoning quality using only a small, general-purpose LLM. CLARity integrates a consistency-aware reward mechanism with a 2-stage refine-then-monitor training pipeline to enhance reasoning consistency, and a dynamic data reformulation strategy to to better exploit limited data. Experiments demonstrate that CLARity improves response consistency by 16.5% and accuracy by 7.5% over baselines. Human evaluations further confirm holistic improvements in coherence and professionalism. Thus, CLARity offers a generalizable solution that enables smaller models to effectively guide expert models by reasoning consistency.Our code is open sourced at: https://github.com/Infinite-set/CLARity

The growing ubiquity of conversational AI highlights the need for frameworks that capture not only users' instrumental goals but also the situated, adaptive, and social practices through which they achieve them. Existing taxonomies of conversational behavior either overgeneralize, remain domain-specific, or reduce interactions to narrow dialogue functions. To address this gap, we introduce the Taxonomy of User Needs and Actions (TUNA), an empirically grounded framework developed through iterative qualitative analysis of 1193 human-AI conversations, supplemented by theoretical review and validation across diverse contexts. TUNA organizes user actions into a three-level hierarchy encompassing behaviors associated with information seeking, synthesis, procedural guidance, content creation, social interaction, and meta-conversation. By centering user agency and appropriation practices, TUNA enables multi-scale evaluation, supports policy harmonization across products, and provides a backbone for layering domain-specific taxonomies. This work contributes a systematic vocabulary for describing AI use, advancing both scholarly understanding and practical design of safer, more responsive, and more accountable conversational systems.