The application of Reinforcement Learning with Verifiable Rewards (RLVR) to mathematical and coding domains has demonstrated significant improvements in the reasoning and problem-solving abilities of Large Language Models. Despite its success in single generation problem solving, the reinforcement learning fine-tuning process may harm the model's exploration ability, as reflected in decreased diversity of generations and a resulting degradation of performance during Best-of-N sampling for large N values. In this work, we focus on optimizing the max@k metric, a continuous generalization of pass@k. We derive an unbiased on-policy gradient estimate for direct optimization of this metric. Furthermore, we extend our derivations to the off-policy updates, a common element in modern RLVR algorithms, that allows better sample efficiency. Empirically, we show that our objective effectively optimizes max@k metric in off-policy scenarios, aligning the model with the Best-of-N inference strategy.

The rapid advancement of large language models (LLMs) has spurred the emergence of data agents--autonomous systems designed to orchestrate Data + AI ecosystems for tackling complex data-related tasks. However, the term "data agent" currently suffers from terminological ambiguity and inconsistent adoption, conflating simple query responders with sophisticated autonomous architectures. This terminological ambiguity fosters mismatched user expectations, accountability challenges, and barriers to industry growth. Inspired by the SAE J3016 standard for driving automation, this survey introduces the first systematic hierarchical taxonomy for data agents, comprising six levels that delineate and trace progressive shifts in autonomy, from manual operations (L0) to a vision of generative, fully autonomous data agents (L5), thereby clarifying capability boundaries and responsibility allocation. Through this lens, we offer a structured review of existing research arranged by increasing autonomy, encompassing specialized data agents for data management, preparation, and analysis, alongside emerging efforts toward versatile, comprehensive systems with enhanced autonomy. We further analyze critical evolutionary leaps and technical gaps for advancing data agents, especially the ongoing L2-to-L3 transition, where data agents evolve from procedural execution to autonomous orchestration. Finally, we conclude with a forward-looking roadmap, envisioning the advent of proactive, generative data agents.

Large Audio Language Models (LALMs), which couple acoustic perception with large language models (LLMs) to extract and understand diverse information from audio, have attracted intense interest from both academic and industrial communities. However, existing LALMs are highly sensitive to how instructions are phrased, affecting both (i) instruction-following rates and (ii) task performance. Yet, no existing benchmarks offer a systematic and comprehensive evaluation of this sensitivity. We introduce ISA-Bench, a dynamic benchmark evaluating instruction sensitivity for LALMs along three axes: instruction description, output format, and task composition. We assess recent open-source and proprietary LALMs using ISA-Bench, profiling both compliance and accuracy under controlled instruction variations. Experimental results reveal that even state-of-the-art LALMs suffer significant instruction sensitivity, leading to degraded performance on fundamental audio understanding tasks. To mitigate this issue, we fine-tune Qwen2-Audio on a specifically constructed complex instruction-variant dataset, achieving a marked improvement in instruction-following performance. However, this also induces nontrivial catastrophic forgetting: the model loses some previously mastered task capabilities when exposed to new instruction styles. Our benchmark provides a standardized basis for assessing and improving instruction sensitivity in LALMs, underscoring the need for instruction-robust audio understanding in real-world pipelines.

Enterprises need access decisions that satisfy least privilege, comply with regulations, and remain auditable. We present a policy aware controller that uses a large language model (LLM) to interpret natural language requests against written policies and metadata, not raw data. The system, implemented with Google Gemini~2.0 Flash, executes a six-stage reasoning framework (context interpretation, user validation, data classification, business purpose test, compliance mapping, and risk synthesis) with early hard policy gates and deny by default. It returns APPROVE, DENY, CONDITIONAL together with cited controls and a machine readable rationale. We evaluate on fourteen canonical cases across seven scenario families using a privacy preserving benchmark. Results show Exact Decision Match improving from 10/14 to 13/14 (92.9\%) after applying policy gates, DENY recall rising to 1.00, False Approval Rate on must-deny families dropping to 0, and Functional Appropriateness and Compliance Adherence at 14/14. Expert ratings of rationale quality are high, and median latency is under one minute. These findings indicate that policy constrained LLM reasoning, combined with explicit gates and audit trails, can translate human readable policies into safe, compliant, and traceable machine decisions.

Recent advances in Visual Language Models (VLMs) have demonstrated exceptional performance in visual reasoning tasks. However, geo-localization presents unique challenges, requiring the extraction of multigranular visual cues from images and their integration with external world knowledge for systematic reasoning. Current approaches to geo-localization tasks often lack robust reasoning mechanisms and explainability, limiting their effectiveness. To address these limitations, we propose the Geo Reason Enhancement (GRE) Suite, a novel framework that augments VLMs with structured reasoning chains for accurate and interpretable location inference. The GRE Suite is systematically developed across three key dimensions: dataset, model, and benchmark. First, we introduce GRE30K, a high-quality geo-localization reasoning dataset designed to facilitate fine-grained visual and contextual analysis. Next, we present the GRE model, which employs a multi-stage reasoning strategy to progressively infer scene attributes, local details, and semantic features, thereby narrowing down potential geographic regions with enhanced precision. Finally, we construct the Geo Reason Evaluation Benchmark (GREval-Bench), a comprehensive evaluation framework that assesses VLMs across diverse urban, natural, and landmark scenes to measure both coarse-grained (e.g., country, continent) and fine-grained (e.g., city, street) localization performance. Experimental results demonstrate that GRE significantly outperforms existing methods across all granularities of geo-localization tasks, underscoring the efficacy of reasoning-augmented VLMs in complex geographic inference. Code and data will be released at https://github.com/Thorin215/GRE.

We propose Fast-MIA (https://github.com/Nikkei/fast-mia), a Python library for efficiently evaluating membership inference attacks (MIA) against Large Language Models (LLMs). MIA against LLMs has emerged as a crucial challenge due to growing concerns over copyright, security, and data privacy, and has attracted increasing research attention. However, the progress of this research is significantly hindered by two main obstacles: (1) the high computational cost of inference in LLMs, and (2) the lack of standardized and maintained implementations of MIA methods, which makes large-scale empirical comparison difficult. To address these challenges, our library provides fast batch inference and includes implementations of representative MIA methods under a unified evaluation framework. This library supports easy implementation of reproducible benchmarks with simple configuration and extensibility. We release Fast-MIA as an open-source (Apache License 2.0) tool to support scalable and transparent research on LLMs.

Generative models have made significant progress in synthesizing high-fidelity audio from short textual descriptions. However, editing existing audio using natural language has remained largely underexplored. Current approaches either require the complete description of the edited audio or are constrained to predefined edit instructions that lack flexibility. In this work, we introduce SAO-Instruct, a model based on Stable Audio Open capable of editing audio clips using any free-form natural language instruction. To train our model, we create a dataset of audio editing triplets (input audio, edit instruction, output audio) using Prompt-to-Prompt, DDPM inversion, and a manual editing pipeline. Although partially trained on synthetic data, our model generalizes well to real in-the-wild audio clips and unseen edit instructions. We demonstrate that SAO-Instruct achieves competitive performance on objective metrics and outperforms other audio editing approaches in a subjective listening study. To encourage future research, we release our code and model weights.

Accurate symptom-to-disease classification and clinically grounded treatment recommendations remain challenging, particularly in heterogeneous patient settings with high diagnostic risk. Existing large language model (LLM)-based systems often lack medical grounding and fail to quantify uncertainty, resulting in unsafe outputs. We propose CLIN-LLM, a safety-constrained hybrid pipeline that integrates multimodal patient encoding, uncertainty-calibrated disease classification, and retrieval-augmented treatment generation. The framework fine-tunes BioBERT on 1,200 clinical cases from the Symptom2Disease dataset and incorporates Focal Loss with Monte Carlo Dropout to enable confidence-aware predictions from free-text symptoms and structured vitals. Low-certainty cases (18%) are automatically flagged for expert review, ensuring human oversight. For treatment generation, CLIN-LLM employs Biomedical Sentence-BERT to retrieve top-k relevant dialogues from the 260,000-sample MedDialog corpus. The retrieved evidence and patient context are fed into a fine-tuned FLAN-T5 model for personalized treatment generation, followed by post-processing with RxNorm for antibiotic stewardship and drug-drug interaction (DDI) screening. CLIN-LLM achieves 98% accuracy and F1 score, outperforming ClinicalBERT by 7.1% (p < 0.001), with 78% top-5 retrieval precision and a clinician-rated validity of 4.2 out of 5. Unsafe antibiotic suggestions are reduced by 67% compared to GPT-5. These results demonstrate CLIN-LLM's robustness, interpretability, and clinical safety alignment. The proposed system provides a deployable, human-in-the-loop decision support framework for resource-limited healthcare environments. Future work includes integrating imaging and lab data, multilingual extensions, and clinical trial validation.

Fact-checking numerical claims is critical as the presence of numbers provide mirage of veracity despite being fake potentially causing catastrophic impacts on society. The prior works in automatic fact verification do not primarily focus on natural numerical claims. A typical human fact-checker first retrieves relevant evidence addressing the different numerical aspects of the claim and then reasons about them to predict the veracity of the claim. Hence, the search process of a human fact-checker is a crucial skill that forms the foundation of the verification process. Emulating a real-world setting is essential to aid in the development of automated methods that encompass such skills. However, existing benchmarks employ heuristic claim decomposition approaches augmented with weakly supervised web search to collect evidences for verifying claims. This sometimes results in less relevant evidences and noisy sources with temporal leakage rendering a less realistic retrieval setting for claim verification. Hence, we introduce QuanTemp++: a dataset consisting of natural numerical claims, an open domain corpus, with the corresponding relevant evidence for each claim. The evidences are collected through a claim decomposition process approximately emulating the approach of human fact-checker and veracity labels ensuring there is no temporal leakage. Given this dataset, we also characterize the retrieval performance of key claim decomposition paradigms. Finally, we observe their effect on the outcome of the verification pipeline and draw insights. The code for data pipeline along with link to data can be found at https://github.com/VenkteshV/QuanTemp_Plus

Large Language Models (LLMs) like ChatGPT and LLaMA drive rapid progress in generative AI, yet their huge parameter scales create severe computational and environmental burdens. High training costs, energy use, and limited device deployment hinder accessibility. Existing compression - pruning, distillation, low-rank, and quantization - reduces size but ignores complex inter-layer correlations. We propose KARIPAP, a quantum-inspired tensor network compression using Infinite Projected Entangled Pair States (iPEPS) and Tensor Renormalization Group (TRG) contraction. Unlike 1D Matrix Product States, iPEPS captures multi-directional entanglement in attention and deep transformer layers. TRG ensures polynomial-time contraction, making tensorization feasible while preserving key correlation geometry. Experiments on LLaMA-2 7B show up to 93% memory and 70% parameter reduction, with 50% faster training, 25% faster inference, and only 2-3% accuracy loss. Layer-wise entanglement profiling reveals redundancy in deeper layers, confirming their suitability for tensor factorization. KARIPAP demonstrates that modern LLMs occupy low-dimensional entanglement manifolds, enabling scalable, energy-efficient, and quantum-aware AI architectures.