Nevertheless, in somesettings,suchassocialnetworks,infectiousdiseasemodeling,andreasoning with spatial and temporal data, this assumption is false.

Understanding the underlying mechanisms is crucial for tasks such asexplaining aphenomenon, predicting, anddecision making. Pearl(2009) providedamachinery for automating the process of answering interventional and (retrospective) counterfactual queries even when only observed data is available, and determining if a query cannot be answered given the available data type (identifiability). This requires knowledge about the true underlying causal structure; however,inmanyreal-world situations, thisstructure isunknown.

Large Language Models (LLMs) have demonstrated impressive capabilities in complex reasoning tasks, yet they still struggle to reliably verify the correctness of their own outputs. Existing solutions to this verification challenge often depend on separate verifier models or require multi-stage self-correction training pipelines, which limit scalability. In this paper, we propose Policy as Generative Verifier (PAG), a simple and effective framework that empowers LLMs to self-correct by alternating between policy and verifier roles within a unified multi-turn reinforcement learning (RL) paradigm. Distinct from prior approaches that always generate a second attempt regardless of model confidence, PAG introduces a selective revision mechanism: the model revises its answer only when its own generative verification step detects an error. This verify-then-revise workflow not only alleviates model collapse but also jointly enhances both reasoning and verification abilities. Extensive experiments across diverse reasoning benchmarks highlight PAG's dual advancements: as a policy, it enhances direct generation and self-correction accuracy; as a verifier, its self-verification outperforms self-consistency.

Currently, large models are prone to generating harmful content when faced with complex attack instructions, significantly reducing their defensive capabilities. To address this issue, this paper proposes a method based on constructing data aligned with multi-dimensional attack defense to enhance the generative security of large models. The core of our method lies in improving the effectiveness of safe alignment learning for large models by innova-tively increasing the diversity of attack instruction dimensions and the accuracy of generat-ing safe responses. To validate the effectiveness of our method, beyond existing security evaluation benchmarks, we additionally designed new security evaluation benchmarks and conducted comparative experiments using Llama3.2 as the baseline model. The final ex-perimental results demonstrate that our method can significantly improve the generative security of large models under complex instructional attacks, while also maintaining and enhancing the models' general capabilities.

We present MMDS, a system capable of recognizing medical images and patient facial details, and providing professional medical diagnoses. The system consists of two core components:The first component is the analysis of medical images and videos. We trained a specialized multimodal medical model capable of interpreting medical images and accurately analyzing patients' facial emotions and facial paralysis conditions. The model achieved an accuracy of 72.59% on the FER2013 facial emotion recognition dataset, with a 91.1% accuracy in recognizing the "happy" emotion. In facial paralysis recognition, the model reached an accuracy of 92%, which is 30% higher than that of GPT-4o. Based on this model, we developed a parser for analyzing facial movement videos of patients with facial paralysis, achieving precise grading of the paralysis severity. In tests on 30 videos of facial paralysis patients, the system demonstrated a grading accuracy of 83.3%.The second component is the generation of professional medical responses. We employed a large language model, integrated with a medical knowledge base, to generate professional diagnoses based on the analysis of medical images or videos. The core innovation lies in our development of a department-specific knowledge base routing management mechanism, in which the large language model categorizes data by medical departments and, during the retrieval process, determines the appropriate knowledge base to query. This significantly improves retrieval accuracy in the RAG (retrieval-augmented generation) process.

This paper delves into the continuous post-training optimization methods for small language models, and proposes a continuous post-training alignment data construction method for small language models. The core of this method is based on the data guidance of large models, optimizing the diversity and accuracy of alignment data. In addition, to verify the effectiveness of the methods in this paper, we used Qwen2-0.5B-Instruct model as the baseline model for small language models, using the alignment dataset constructed by our proposed method, we trained and compared several groups of experiments, including SFT (Supervised Fine Tuning) post-training experiment and KTO (Kahneman Tversky optimization) post-training experiment, as well as SFT-KTO two-stage post-training experiment and model weight fusion experiment. Finally, we evaluated and analyzed the performance of post-training models, and confirmed that the continuous post-training optimization method proposed by us can significantly improve the performance of small language models.

Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). These techniques are often not applicable in unconditional generation or in various downstream tasks such as image restoration. In this paper, we propose a novel sampling guidance, called Perturbed-Attention Guidance (PAG), which improves diffusion sample quality across both unconditional and conditional settings, achieving this without requiring additional training or the integration of external modules. PAG is designed to progressively enhance the structure of samples throughout the denoising process. It involves generating intermediate samples with degraded structure by substituting selected self-attention maps in diffusion U-Net with an identity matrix, by considering the self-attention mechanisms' ability to capture structural information, and guiding the denoising process away from these degraded samples. In both ADM and Stable Diffusion, PAG surprisingly improves sample quality in conditional and even unconditional scenarios. Moreover, PAG significantly improves the baseline performance in various downstream tasks where existing guidances such as CG or CFG cannot be fully utilized, including ControlNet with empty prompts and image restoration such as inpainting and deblurring.

This paper addresses the problem of integrating local guide policies into a Reinforcement Learning agent. For this, we show how to adapt existing algorithms to this setting before introducing a novel algorithm based on a noisy policy-switching procedure. This approach builds on a proper Approximate Policy Evaluation (APE) scheme to provide a perturbation that carefully leads the local guides towards better actions. We evaluated our method on a set of classical Reinforcement Learning problems, including safety-critical systems where the agent cannot enter some areas at the risk of triggering catastrophic consequences. In all the proposed environments, our agent proved to be efficient at leveraging those policies to improve the performance of any APE-based Reinforcement Learning algorithm, especially in its first learning stages.

Causal discovery is an essential part of causal inference (Spirtes et al., 2000; Peters et al., 2017), but estimating causal effects is extremely challenging if the underlying causal graph is unknown. Algorithms for learning causal graphs are many and varied, using different parametric structure, classes of graphical models, and assumptions about whether all relevant variables are measured (Spirtes et al., 2000; Kaltenpoth and Vreeken, 2023; Claassen and Bucur, 2022; Nowzohour et al., 2017; Zhang and Hyvarinen, 2009; Peters et al., 2017). In this paper, we consider only nonparametric assumptions, i.e. conditional independences in distributions that are represented by graphs. The primary graphical model used in causal inference is the directed acyclic graph, also known as a DAG. These offer a clear interpretation and are straightforward to conduct inference with, and are associated with probabilistic distributions by encoding conditional independence constraints.

Predicting the effect of unseen interventions is a fundamental research question across the data sciences. It is well established that, in general, such questions cannot be answered definitively from observational data, e.g., as a consequence of unobserved confounding. A generalization of this task is to determine non-trivial bounds on causal effects induced by the data, also known as the task of partial causal identification. In the literature, several algorithms have been developed for solving this problem. Most, however, require a known parametric form or a fully specified causal diagram as input, which is usually not available in practical applications. In this paper, we assume as input a less informative structure known as a Partial Ancestral Graph, which represents a Markov equivalence class of causal diagrams and is learnable from observational data. In this more "data-driven" setting, we provide a systematic algorithm to derive bounds on causal effects that can be computed analytically.