Large language models (LLMs) excel at generating long-form responses, but evaluating their factuality remains challenging due to complex inter-sentence dependencies within the generated facts. Prior solutions predominantly follow a decompose-decontextualize-verify pipeline but often fail to capture essential context and miss key relational facts. In this paper, we introduce VeriFact, a factuality evaluation framework designed to enhance fact extraction by identifying and resolving incomplete and missing facts to support more accurate verification results. Moreover, we introduce FactRBench , a benchmark that evaluates both precision and recall in long-form model responses, whereas prior work primarily focuses on precision. FactRBench provides reference fact sets from advanced LLMs and human-written answers, enabling recall assessment. Empirical evaluations show that VeriFact significantly enhances fact completeness and preserves complex facts with critical relational information, resulting in more accurate factuality evaluation. Benchmarking various open- and close-weight LLMs on FactRBench indicate that larger models within same model family improve precision and recall, but high precision does not always correlate with high recall, underscoring the importance of comprehensive factuality assessment.
Large Language Models (LLMs) frequently produce factoid hallucinations - plausible yet incorrect answers. A common mitigation strategy is model alignment, which improves factual accuracy by training on curated factual and non-factual pairs. However, this approach often relies on a stronger model (e.g., GPT-4) or an external knowledge base to assess factual correctness, which may not always be accessible. To address this, we propose Atomic Consistency Preference Optimization (ACPO), a self-supervised preference-tuning method that enhances factual accuracy without external supervision. ACPO leverages atomic consistency signals, i.e., the agreement of individual facts across multiple stochastic responses, to identify high- and low-quality data pairs for model alignment. By eliminating the need for costly GPT calls, ACPO provides a scalable and efficient approach to improving factoid question-answering. Despite being self-supervised, empirical results demonstrate that ACPO outperforms FactAlign, a strong supervised alignment baseline, by 1.95 points on the LongFact and BioGen datasets, highlighting its effectiveness in enhancing factual reliability without relying on external models or knowledge bases.
We employ a unified framework for computing the information capacity of biological signaling systems using Fisher Information. By deriving closed-form or easily computable information capacity formulas, we quantify how well different signaling models, including binomial, multinomial, Poisson, Gaussian, and Gamma distributions, can discriminate among input signals. These expressions clarify how key features such as signal range, noise scaling, pathway length, and receivers' diversity shape the theoretical limits of sensing. In particular, we show how signal-to-noise ratio and fold-change sensitivity arise naturally within the Fisher formalism, and how signal degradation in cascades imposes linear information loss. Our results provide intuitive, analytically grounded tools to benchmark and guide the analysis of real signaling systems, without requiring computationally expensive mutual information estimation. While motivated by cellular communication, the framework generalizes to any system where noisy input-output relationships constrain transmission fidelity, including synthetic biology, sensor networks, and engineered communication channels.
Despite recent advances in video understanding, the capabilities of Large Video Language Models (LVLMs) to perform video-based causal reasoning remains underexplored, largely due to the absence of relevant and dedicated benchmarks for evaluating causal reasoning in visually grounded and goal-driven settings. To fill this gap, we introduce a novel benchmark named Video-based long-form Causal Reasoning (VCRBench). We create VCRBench using procedural videos of simple everyday activities, where the steps are deliberately shuffled with each clip capturing a key causal event, to test whether LVLMs can identify, reason about, and correctly sequence the events needed to accomplish a specific goal. Moreover, the benchmark is carefully designed to prevent LVLMs from exploiting linguistic shortcuts, as seen in multiple-choice or binary QA formats, while also avoiding the challenges associated with evaluating open-ended QA. Our evaluation of state-of-the-art LVLMs on VCRBench suggests that these models struggle with video-based long-form causal reasoning, primarily due to their difficulty in modeling long-range causal dependencies directly from visual observations. As a simple step toward enabling such capabilities, we propose Recognition-Reasoning Decomposition (RRD), a modular approach that breaks video-based causal reasoning into two sub-tasks of video recognition and causal reasoning. Our experiments on VCRBench show that RRD significantly boosts accuracy on VCRBench, with gains of up to 25.2%. Finally, our thorough analysis reveals interesting insights, for instance, that LVLMs primarily rely on language knowledge for complex video-based long-form causal reasoning tasks.
This paper introduces StoryAnchors, a unified framework for generating high-quality, multi-scene story frames with strong temporal consistency. The framework employs a bidirectional story generator that integrates both past and future contexts to ensure temporal consistency, character continuity, and smooth scene transitions throughout the narrative. Specific conditions are introduced to distinguish story frame generation from standard video synthesis, facilitating greater scene diversity and enhancing narrative richness. To further improve generation quality, StoryAnchors integrates Multi-Event Story Frame Labeling and Progressive Story Frame Training, enabling the model to capture both overarching narrative flow and event-level dynamics. This approach supports the creation of editable and expandable story frames, allowing for manual modifications and the generation of longer, more complex sequences. Extensive experiments show that StoryAnchors outperforms existing open-source models in key areas such as consistency, narrative coherence, and scene diversity. Its performance in narrative consistency and story richness is also on par with GPT-4o. Ultimately, StoryAnchors pushes the boundaries of story-driven frame generation, offering a scalable, flexible, and highly editable foundation for future research.
As Large Language Models (LLMs) become widely accessible, a detailed understanding of their knowledge within specific domains becomes necessary for successful real world use. This is particularly critical in public health, where failure to retrieve relevant, accurate, and current information could significantly impact UK residents. However, currently little is known about LLM knowledge of UK Government public health information. To address this issue, this paper introduces a new benchmark, PubHealthBench, with over 8000 questions for evaluating LLMs' Multiple Choice Question Answering (MCQA) and free form responses to public health queries. To create PubHealthBench we extract free text from 687 current UK government guidance documents and implement an automated pipeline for generating MCQA samples. Assessing 24 LLMs on PubHealthBench we find the latest private LLMs (GPT-4.5, GPT-4.1 and o1) have a high degree of knowledge, achieving >90% accuracy in the MCQA setup, and outperform humans with cursory search engine use. However, in the free form setup we see lower performance with no model scoring >75%. Importantly we find in both setups LLMs have higher accuracy on guidance intended for the general public. Therefore, there are promising signs that state of the art (SOTA) LLMs are an increasingly accurate source of public health information, but additional safeguards or tools may still be needed when providing free form responses on public health topics.
Recently, zero-shot anomaly detection (ZSAD) has emerged as a pivotal paradigm for identifying defects in unseen categories without requiring target samples in training phase. However, existing ZSAD methods struggle with the boundary of small and complex defects due to insufficient representations. Most of them use the single manually designed prompts, failing to work for diverse objects and anomalies. In this paper, we propose MFP-CLIP, a novel prompt-based CLIP framework which explores the efficacy of multi-form prompts for zero-shot industrial anomaly detection. We employ an image to text prompting(I2TP) mechanism to better represent the object in the image. MFP-CLIP enhances perception to multi-scale and complex anomalies by self prompting(SP) and a multi-patch feature aggregation(MPFA) module. To precisely localize defects, we introduce the mask prompting(MP) module to guide model to focus on potential anomaly regions. Extensive experiments are conducted on two wildly used industrial anomaly detection benchmarks, MVTecAD and VisA, demonstrating MFP-CLIP's superiority in ZSAD.
We show that assuming that the returns are independent when conditioned on the value of their variance (volatility), which itself varies in time randomly, then the distribution of returns is well described by the statistics of the sum of conditionally independent random variables. In particular, we show that the distribution of returns can be cast in a simple scaling form, and that its functional form is directly related to the distribution of the volatilities. This approach explains the presence of power-law tails in the returns as a direct consequence of the presence of a power law tail in the distribution of volatilities. It also provides the form of the distribution of Bitcoin returns, which behaves as a stretched exponential, as a consequence of the fact that the Bitcoin volatilities distribution is also closely described by a stretched exponential. We test our predictions with data from the S\&P 500 index, Apple and Paramount stocks; and Bitcoin.
A solution concept that is a refinement of Nash equilibria selects for each finite game a nonempty collection of closed and connected subsets of Nash equilibria as solutions. We impose three axioms for such solution concepts. The axiom of backward induction requires each solution to contain a quasi-perfect equilibrium. Two invariance axioms posit that solutions of a game are the same as those of a game obtained by the addition of strategically irrelevant strategies and players. Stability satisfies these axioms; and any solution concept that satisfies them must, for generic extensive-form games, select from among its stable outcomes. A strengthening of the two invariance axioms provides an analogous axiomatization of components of equilibria with a nonzero index.
Standard procurement models assume that the buyer knows the quality of the good at the time of procurement; however, in many settings, the quality is learned only long after the transaction. We study procurement problems in which the buyer's valuation of the supplied good depends directly on its quality, which is unverifiable and unobservable to the buyer. For a broad class of procurement problems, we identify procurement mechanisms maximizing any weighted average of the buyer's expected payoff and social surplus. The optimal mechanism can be implemented by an auction that restricts sellers to submit bids within specific intervals.
Extraction tests of Compton Form Factors are performed using pseudodata based on experimental data from Deeply Virtual Compton Scattering experiments conducted at Jefferson Lab. The standard Belitsky, Kirchner, and Muller formalism at twist-two is employed, along with a fitting procedure designed to reduce model dependency similar to traditional local fits. The extraction of the Compton Form Factors is performed using both Classical Deep Neural Networks (CDNNs) and Quantum Deep Neural Networks (QDNNs). Comparative studies reveal that QDNNs outperform CDNNs for this application, demonstrating improved predictive accuracy and precision even for limited model complexity. The results demonstrate the potential of QDNNs for future studies in which quantum algorithms can be fully optimized.
The development of powerful user representations is a key factor in the success of recommender systems (RecSys). Online platforms employ a range of RecSys techniques to personalize user experience across diverse in-app surfaces. User representations are often learned individually through user's historical interactions within each surface and user representations across different surfaces can be shared post-hoc as auxiliary features or additional retrieval sources. While effective, such schemes cannot directly encode collaborative filtering signals across different surfaces, hindering its capacity to discover complex relationships between user behaviors and preferences across the whole platform. To bridge this gap at Snapchat, we seek to conduct universal user modeling (UUM) across different in-app surfaces, learning general-purpose user representations which encode behaviors across surfaces. Instead of replacing domain-specific representations, UUM representations capture cross-domain trends, enriching existing representations with complementary information. This work discusses our efforts in developing initial UUM versions, practical challenges, technical choices and modeling and research directions with promising offline performance. Following successful A/B testing, UUM representations have been launched in production, powering multiple use cases and demonstrating their value. UUM embedding has been incorporated into (i) Long-form Video embedding-based retrieval, leading to 2.78% increase in Long-form Video Open Rate, (ii) Long-form Video L2 ranking, with 19.2% increase in Long-form Video View Time sum, (iii) Lens L2 ranking, leading to 1.76% increase in Lens play time, and (iv) Notification L2 ranking, with 0.87% increase in Notification Open Rate.
\Ac{LFQA} aims to generate lengthy answers to complex questions. This scenario presents great flexibility as well as significant challenges for evaluation. Most evaluations rely on deterministic metrics that depend on string or n-gram matching, while the reliability of large language model-based evaluations for long-form answers remains relatively unexplored. We address this gap by conducting an in-depth study of long-form answer evaluation with the following research questions: (i) To what extent do existing automatic evaluation metrics serve as a substitute for human evaluations? (ii) What are the limitations of existing evaluation metrics compared to human evaluations? (iii) How can the effectiveness and robustness of existing evaluation methods be improved? We collect 5,236 factoid and non-factoid long-form answers generated by different large language models and conduct a human evaluation on 2,079 of them, focusing on correctness and informativeness. Subsequently, we investigated the performance of automatic evaluation metrics by evaluating these answers, analyzing the consistency between these metrics and human evaluations. We find that the style, length of the answers, and the category of questions can bias the automatic evaluation metrics. However, fine-grained evaluation helps mitigate this issue on some metrics. Our findings have important implications for the use of large language models for evaluating long-form question answering. All code and datasets are available at https://github.com/bugtig6351/lfqa_evaluation.
Kernel mean embeddings -- integrals of a kernel with respect to a probability distribution -- are essential in Bayesian quadrature, but also widely used in other computational tools for numerical integration or for statistical inference based on the maximum mean discrepancy. These methods often require, or are enhanced by, the availability of a closed-form expression for the kernel mean embedding. However, deriving such expressions can be challenging, limiting the applicability of kernel-based techniques when practitioners do not have access to a closed-form embedding. This paper addresses this limitation by providing a comprehensive dictionary of known kernel mean embeddings, along with practical tools for deriving new embeddings from known ones. We also provide a Python library that includes minimal implementations of the embeddings.
Tasks for multi-robot systems often require the robots to collaborate and complete a team goal while maintaining safety. This problem is usually formalized as a constrained Markov decision process (CMDP), which targets minimizing a global cost and bringing the mean of constraint violation below a user-defined threshold. Inspired by real-world robotic applications, we define safety as zero constraint violation. While many safe multi-agent reinforcement learning (MARL) algorithms have been proposed to solve CMDPs, these algorithms suffer from unstable training in this setting. To tackle this, we use the epigraph form for constrained optimization to improve training stability and prove that the centralized epigraph form problem can be solved in a distributed fashion by each agent. This results in a novel centralized training distributed execution MARL algorithm named Def-MARL. Simulation experiments on 8 different tasks across 2 different simulators show that Def-MARL achieves the best overall performance, satisfies safety constraints, and maintains stable training. Real-world hardware experiments on Crazyflie quadcopters demonstrate the ability of Def-MARL to safely coordinate agents to complete complex collaborative tasks compared to other methods.
Developing large audio language models (LMs) capable of understanding diverse spoken interactions is essential for accommodating the multimodal nature of human communication and can increase the accessibility of language technologies across different user populations. Recent work on audio LMs has primarily evaluated their performance on short audio segments, typically under 30 seconds, with limited exploration of long-form conversational speech segments that more closely reflect natural user interactions with these models. We introduce Brutally Long Audio Bench (BLAB), a challenging long-form audio benchmark that evaluates audio LMs on localization, duration estimation, emotion, and counting tasks using audio segments averaging 51 minutes in length. BLAB consists of 833+ hours of diverse, full-length audio clips, each paired with human-annotated, text-based natural language questions and answers. Our audio data were collected from permissively licensed sources and underwent a human-assisted filtering process to ensure task compliance. We evaluate six open-source and proprietary audio LMs on BLAB and find that all of them, including advanced models such as Gemini 2.0 Pro and GPT-4o, struggle with the tasks in BLAB. Our comprehensive analysis reveals key insights into the trade-offs between task difficulty and audio duration. In general, we find that audio LMs struggle with long-form speech, with performance declining as duration increases. They perform poorly on localization, temporal reasoning, counting, and struggle to understand non-phonemic information, relying more on prompts than audio content. BLAB serves as a challenging evaluation framework to develop audio LMs with robust long-form audio understanding capabilities.
Artificial Kuramoto oscillatory neurons were recently introduced as an alternative to threshold units. Empirical evidence suggests that oscillatory units outperform threshold units in several tasks including unsupervised object discovery and certain reasoning problems. The proposed coupling mechanism for these oscillatory neurons is heterogeneous, combining a generalized Kuramoto equation with standard coupling methods used for threshold units. In this research note, we present a theoretical framework that clearly distinguishes oscillatory neurons from threshold units and establishes a coupling mechanism between them. We argue that, from a biological standpoint, oscillatory and threshold units realise distinct aspects of neural coding: roughly, threshold units model intensity of neuron firing, while oscillatory units facilitate information exchange by frequency modulation. To derive interaction between these two types of units, we constrain their dynamics by focusing on dynamical systems that admit Lyapunov functions. For threshold units, this leads to Hopfield associative memory model, and for oscillatory units it yields a specific form of generalized Kuramoto model. The resulting dynamical systems can be naturally coupled to form a Hopfield-Kuramoto associative memory model, which also admits a Lyapunov function. Various forms of coupling are possible. Notably, oscillatory neurons can be employed to implement a low-rank correction to the weight matrix of a Hopfield network. This correction can be viewed either as a form of Hebbian learning or as a popular LoRA method used for fine-tuning of large language models. We demonstrate the practical realization of this particular coupling through illustrative toy experiments.
Internet telephony consists of a combination of hardware and software that enables you to use the Internet as the transmission medium for telephone calls. For users who have free, or fixed-price Internet access, Internet telephony software essentially provides free telephone calls anywhere in the world. In its simplest form, PC-to-PC Internet telephony can be as easy as hooking up a microphone to your computer and sending your voice through a cable modem to a person who has Internet telephony software that is compatible with yours. This basic form of Internet telephony is not without its problems, however. Connecting this way is slower than using a traditional telephone, and the quality of the voice transmissions is also not near the quality you would get when placing a regular phone call.
In this work, we build the first benchmark dataset for short-form UGC Image Super-resolution in the wild, termed KwaiSR, intending to advance the research on developing image super-resolution algorithms for short-form UGC platforms. This dataset is collected from the Kwai Platform, which is composed of two parts, i.e., synthetic and wild parts. Among them, the synthetic dataset, including 1,900 image pairs, is produced by simulating the degradation following the distribution of real-world low-quality short-form UGC images, aiming to provide the ground truth for training and objective comparison in the validation/testing. The wild dataset contains low-quality images collected directly from the Kwai Platform, which are filtered using the quality assessment method KVQ from the Kwai Platform. As a result, the KwaiSR dataset contains 1800 synthetic image pairs and 1900 wild images, which are divided into training, validation, and testing parts with a ratio of 8:1:1. Based on the KwaiSR dataset, we organize the NTIRE 2025 challenge on a second short-form UGC Video quality assessment and enhancement, which attracts lots of researchers to develop the algorithm for it. The results of this competition have revealed that our KwaiSR dataset is pretty challenging for existing Image SR methods, which is expected to lead to a new direction in the image super-resolution field. The dataset can be found from https://lixinustc.github.io/NTIRE2025-KVQE-KwaSR-KVQ.github.io/.
Process reward models (PRMs) have proven effective for test-time scaling of Large Language Models (LLMs) on challenging reasoning tasks. However, reward hacking issues with PRMs limit their successful application in reinforcement fine-tuning. In this paper, we identify the main cause of PRM-induced reward hacking: the canonical summation-form credit assignment in reinforcement learning (RL), which defines the value as cumulative gamma-decayed future rewards, easily induces LLMs to hack steps with high rewards. To address this, we propose PURE: Process sUpervised Reinforcement lEarning. The key innovation of PURE is a min-form credit assignment that formulates the value function as the minimum of future rewards. This method significantly alleviates reward hacking by limiting the value function range and distributing advantages more reasonably. Through extensive experiments on 3 base models, we show that PRM-based approaches enabling min-form credit assignment achieve comparable reasoning performance to verifiable reward-based methods within only 30% steps. In contrast, the canonical sum-form credit assignment collapses training even at the beginning! Additionally, when we supplement PRM-based fine-tuning with just 10% verifiable rewards, we further alleviate reward hacking and produce the best fine-tuned model based on Qwen2.5-Math-7B in our experiments, achieving 82.5% accuracy on AMC23 and 53.3% average accuracy across 5 benchmarks. Moreover, we summarize the observed reward hacking cases and analyze the causes of training collapse. Code and models are available at https://github.com/CJReinforce/PURE.