NeurIPS 2025 Papers

Retrieval-augmented generation (RAG) greatly enhances large language models (LLMs) performance in knowledge-intensive tasks. However, naive RAG methods struggle with multi-hop question answering due to their limited capacity to capture complex dependencies across documents. Recent studies employ graph-based RAG to capture document connections. However, these approaches often result in a loss of semantic coherence and introduce irrelevant noise during node matching and subgraph construction. To address these limitations, we propose NeuroPath, an LLM-driven semantic path tracking RAG framework inspired by the path navigational planning of place cells in neurobiology. It consists of two steps: Dynamic Path Tracking and Post-retrieval Completion. Dynamic Path Tracking performs goal-directed semantic path tracking and pruning over the constructed knowledge graph (KG), improving noise reduction and semantic coherence. Post-retrieval Completion further reinforces these benefits by conducting second-stage retrieval using intermediate reasoning and the original query to refine the query goal and complete missing information in the reasoning path. NeuroPath surpasses current state-of-the-art baselines on three multi-hop QA datasets, achieving average improvements of 16.3\% on recall@2 and 13.5\% on recall@5 over advanced graph-based RAG methods. Moreover, compared to existing iter-based RAG methods, NeuroPath achieves higher accuracy and reduces token consumption by 22.8\%. Finally, we demonstrate the robustness of NeuroPath across four smaller LLMs (Llama3.1, GLM4, Mistral0.3, and Gemma3), and further validate its scalability across tasks of varying complexity. Code is available at [https://github.com/KennyCaty/NeuroPath](https://github.com/KennyCaty/NeuroPath).

We study the problem of approximating all-pair distances in a weighted undirected graph with differential privacy, introduced by Sealfon [Sea16]. Given a publicly known undirected graph, we treat the weights of edges as sensitive information, and two graphs are neighbors if their edge weights differ in one edge by at most one. We obtain efficient algorithms with significantly improved bounds on a broad class of graphs which we refer to as *recursively separable*. In particular, for any $n$-vertex $K_h$-minor-free graph, our algorithm achieve an additive error of $ \widetilde{O}(h(nW)^{1/3} ) $, where $ W $ represents the maximum edge weight; For grid graphs, the same algorithmic scheme achieve additive error of $ \widetilde{O}(n^{1/4}\sqrt{W}) $. Our approach can be seen as a generalization of the celebrated binary tree mechanism for range queries, as releasing range queries is equivalent to computing all-pair distances on a path graph. In essence, our approach is based on generalizing the binary tree mechanism to graphs that are *recursively separable*.

This work introduces Clipout, a method for removing a target concept in pre-trained text-to-image models. By randomly clipping units from the learned data embedding and using a contrastive objective, models are encouraged to differentiate these clipped embedding vectors. Our goal is to remove private, copyrighted, inaccurate, or harmful concepts from trained models without the need for retraining. This is achieved by considering only negative samples and generating them in a bootstrapping-like manner, requiring minimal prior knowledge. Additionally, theoretical analyses are provided to further understand our proposed Clipout. Extensive experiments on text-to-image show that Clipout is simple yet highly effective and efficient compared with previous state-of-the-art approaches.

Although diffusion-based models can generate high-quality and high-resolution video sequences from textual or image inputs, they lack explicit integration of geometric cues when controlling scene lighting and visual appearance across frames. To address this limitation, we propose IllumiCraft, an end-to-end diffusion framework accepting three complementary inputs: (1) high-dynamic-range (HDR) video maps for detailed lighting control; (2) synthetically relit frames with randomized illumination changes (optionally paired with a static background reference image) to provide appearance cues; and (3) 3D point tracks that capture precise 3D geometry information. By integrating the lighting, appearance, and geometry cues within a unified diffusion architecture, IllumiCraft generates temporally coherent videos aligned with user-defined prompts. It supports the background-conditioned and text-conditioned video relighting and provides better fidelity than existing controllable video generation methods.

Detecting AI-involved text is essential for combating misinformation, plagiarism, and academic misconduct. However, AI text generation includes diverse collaborative processes (AI-written text edited by humans, human-written text edited by AI, and AI-generated text refined by other AI), where various or even new LLMs could be involved. Texts generated through these varied processes exhibit complex characteristics, presenting significant challenges for detection. Current methods model these processes rather crudely, primarily employing binary classification (purely human vs. AI-involved) or multi-classification (treating human-AI collaboration as a new class). We observe that representations of texts generated through different processes exhibit inherent clustering relationships. Therefore, we propose DETree, a novel approach that models the relationships among different processes as a Hierarchical Affinity Tree structure, and introduces a specialized loss function that aligns text representations with this tree. To facilitate this learning, we developed RealBench, a comprehensive benchmark dataset that automatically incorporates a wide spectrum of hybrid texts produced through various human-AI collaboration processes. Our method improves performance in hybrid text detection tasks and significantly enhances robustness and generalization in out-of-distribution scenarios, particularly in few-shot learning conditions, further demonstrating the promise of training-based approaches in OOD settings. Our code and dataset are available at https://github.com/heyongxin233/DETree.

The increasing societal integration of Large Language Models (LLMs) and agent-based AI demands a new civic competency: adversarial reasoning. This position paper argues that K-12 AI education must move beyond passive literacy to actively equip students with skills in responsible adversarial prompting and ethical system "hacking." Such capabilities are essential for citizens to critically probe AI systems, understand their inherent limitations, identify manipulative patterns, and hold them accountable. We posit that cultivating a generation skilled in "red-teaming" AI is vital for maintaining transparency, preventing undue influence, and fostering a democratic engagement with these transformative technologies.

Adapting pretrained diffusion-based generative models for text-driven image editing with negligible tuning overhead has demonstrated remarkable potential. A classical adaptation paradigm, as followed by these methods, first infers the generative trajectory inversely for a given source image by image inversion, then performs image editing along the inferred trajectory guided by the target text prompts. However, the performance of image editing is heavily limited by the approximation errors introduced during image inversion by diffusion models, which arise from the absence of exact supervision in the intermediate generative steps. To circumvent this issue, we investigate the parameter-efficient adaptation of VQ-based generative models for image editing, and leverage their inherent characteristic that the exact intermediate quantized representations of a source image are attainable, enabling more effective supervision for precise image inversion. Specifically, we propose \emph{EditInfinity}, which adapts \emph{Infinity}, a binary-quantized generative model, for image editing. We propose an efficient yet effective image inversion mechanism that integrates text prompting rectification and image style preservation, enabling precise image inversion. Furthermore, we devise a holistic smoothing strategy which allows our \emph{EditInfinity} to perform image editing with high fidelity to source images and precise semantic alignment to the text prompts. Extensive experiments on the PIE-Bench benchmark across "add", "change", and "delete" editing operations, demonstrate the superior performance of our model compared to state-of-the-art diffusion-based baselines. Code available at: \url{https://github.com/yx-chen-ust/EditInfinity}.

In this paper, we introduce Online Multimodal Conversational Response Generation (OMCRG), a novel task designed to produce synchronized verbal and non-verbal listener feedback online, based on the speaker's multimodal inputs. OMCRG captures natural dyadic interactions and introduces new challenges in aligning generated audio with listeners' facial responses. To tackle these challenges, we incorporate text as an intermediate modality to connect audio and facial responses. We propose OmniResponse, a Multimodal Large Language Model (MLLM) that autoregressively generates accurate multimodal listener responses. OmniResponse leverages a pretrained LLM enhanced with two core components: Chrono-Text Markup, which precisely timestamps generated text tokens, and TempoVoice, a controllable online text-to-speech (TTS) module that outputs speech synchronized with facial responses. To advance OMCRG research, we offer ResponseNet, a dataset of 696 detailed dyadic interactions featuring synchronized split-screen videos, multichannel audio, transcripts, and annotated facial behaviors. Comprehensive evaluations on ResponseNet demonstrate that OmniResponse outperforms baseline models in terms of semantic speech content, audio-visual synchronization, and generation quality. Our dataset, code, and models are publicly available at https://omniresponse.github.io/.

Recent advances in multi-modal models have demonstrated strong performance in tasks such as image generation and reasoning. However, applying these models to the fire domain remains challenging due to the lack of publicly available datasets with high-quality fire domain annotations. To address this gap, we introduce $\textbf{DetectiumFire}$, a large-scale, multi-modal dataset comprising of 22.5k high-resolution fire-related images and 2.5k real-world fire-related videos covering a wide range of fire types, environments, and risk levels. The data are annotated with both traditional computer vision labels (e.g., bounding boxes) and detailed textual prompts describing the scene, enabling applications such as synthetic data generation and fire risk reasoning. DetectiumFire offers clear advantages over existing benchmarks in scale, diversity, and data quality, significantly reducing redundancy and enhancing coverage of real-world scenarios. We validate the utility of DetectiumFire across multiple tasks, including object detection, diffusion-based image generation, and vision-language reasoning. Our results highlight the potential of this dataset to advance fire-related research and support the development of intelligent safety systems. We release DetectiumFire to promote broader exploration of fire understanding in the AI community.

Times series foundation models (TSFMs) have emerged as a promising paradigm for time series analysis and forecasting, showing remarkable generalization performance across different domains. While efforts have been made on hallucinations of foundation models, the hallucinations of TSFMs have been underexplored. In this paper, we formally define TSFM hallucinations in the zero-shot forecasting setting by examining whether a generated forecast exhibits different dynamics from those of the context. Our study reveals that TSFM hallucinations primarily stem from the loss of context information in hidden states during forward propagation. As such, we propose methodologies to identify signal subspaces in TSFMs and magnify their information through intervention. Extensive experiments demonstrate that our proposed intervention approach effectively mitigates hallucinations and improves forecasting performance. Furthermore, the signal strength measure we compute from signal subspaces has strong predictive power of hallucinations and forecasting performance of the model. Our work contributes to deeper understanding of TSFM trustworthiness that could foster future research in this direction.

Sampling efficiently from a target unnormalized probability density remains a core challenge, with relevance across countless high-impact scientific applications. A promising approach towards this challenge is the design of amortized samplers that borrow key ideas, such as probability path design, from state-of-the-art generative diffusion models. However, all existing diffusion-based samplers remain unable to draw samples from distributions at the scale of even simple molecular systems. In this paper, we propose Progressive Inference-Time Annealing (PITA) a novel framework to learn diffusion-based samplers that combines two complementary interpolation techniques: I.) Annealing of the Boltzmann distribution and II.) Diffusion smoothing. PITA trains a sequence of diffusion models from high to low temperatures by sequentially training each model at progressively higher temperatures, leveraging engineered easy access to samples of the temperature-annealed target density. In the subsequent step, PITA enables simulating the trained diffusion model to *procure training samples at a lower temperature* for the next diffusion model through inference-time annealing using a novel Feynman-Kac PDE combined with Sequential Monte Carlo. Empirically, PITA enables, for the first time, equilibrium sampling of $N$-body particle systems, Alanine Dipeptide, and tripeptides in Cartesian coordinates with dramatically lower energy function evaluations.

Model-based reinforcement learning (MBRL) hinges on a learned dynamics model whose errors can compound along imagined rollouts. We study how encouraging \emph{flatness} in the model’s training loss affects downstream control, and show that steering optimization toward flatter minima yields a better policy. Concretely, we integrate \emph{Sharpness-Aware Minimization} (SAM) into world-model training as a drop-in objective, leaving the planner and policy components unchanged. On the theory side, we derive PAC-Bayesian bounds that link first-order sharpness to the value-estimation gap and the performance gap between model-optimal and true-optimal policies, implying that flatter minima tighten both. Empirically, SAM reduces measured sharpness and value-prediction error and improves returns across HumanoidBench, Atari-100k, and high-DoF DeepMind Control tasks. Augmenting existing MBRL algorithms with SAM increases mean return, with especially large gains in settings with high dimensional state–action space. We further observe positive transfer across algorithms and input modalities, including a transformer-based world-model. These results position flat-minima training as a simple, general mechanism for more robust MBRL without architectural changes.

Existing benchmarks for evaluating the security risks and capabilities (e.g., vulnerability detection) of code-generating large language models (LLMs) face several key limitations: (1) limited coverage of risk and capabilities; (2) reliance on static evaluation metrics such as LLM judgments or rule-based detection, which lack the precision of dynamic analysis; and (3) a trade-off between data quality and benchmark scale. To address these challenges, we introduce a general and scalable benchmark construction framework that begins with manually validated, high-quality seed examples and expands them via targeted mutations. Each mutated sample retains the seed’s security semantics while providing diverse, unseen instances. The resulting benchmark bundles every artifact required for dynamic evaluation, including prompts, vulnerable and patched code, test cases, and ground-truth proofs of concept, enabling rigorous measurement of insecure coding, vulnerability detection, and patch generation. Applying this framework to Python, C/C++, and Java, we build SECODEPLT, a dataset of more than 5.9k samples spanning 44 CWE-based risk categories and three security capabilities. Compared with state-of-the-art benchmarks, SECODEPLT offers broader coverage, higher data fidelity, and substantially greater scale. We use SECODEPLT to evaluate leading code-generation LLMs and agents, revealing their strengths and weaknesses in both generating secure code and identifying or fixing vulnerabilities. We provide our code in \url{https://github.com/ucsb-mlsec/SeCodePLT}, data in \url{https://huggingface.co/datasets/UCSB-SURFI/SeCodePLT}

Performativity of predictions refers to the phenomenon where prediction-informed decisions influence the very targets they aim to predict—a dynamic commonly observed in policy-making, social sciences, and economics. In this paper, we initiate an end-to-end framework of statistical inference under performativity. Our contributions are twofold. First, we establish a central limit theorem for estimation and inference in the performative setting, enabling standard inferential tasks such as constructing confidence intervals and conducting hypothesis tests in policy-making contexts. Second, we leverage this central limit theorem to study prediction-powered inference (PPI) under performativity. This approach yields more precise estimates and tighter confidence regions for the model parameters (i.e., policies) of interest in performative prediction. We validate the effectiveness of our framework through numerical experiments. To the best of our knowledge, this is the first work to establish a complete statistical inference under performativity, introducing new challenges and inference settings that we believe will provide substantial value to policy-making, statistics, and machine learning.

The recent development of zero-shot reinforcement learning (RL) has opened a new avenue for learning pre-trained generalist policies that can adapt to arbitrary new tasks in a zero-shot manner. While the popular Forward-Backward representations (FB) and related methods have shown promise in zero-shot RL, we empirically found that their modeling lacks expressivity and that extrapolation errors caused by out-of-distribution (OOD) actions during offline learning sometimes lead to biased representations, ultimately resulting in suboptimal performance. To address these issues, we propose Behavior-REgularizEd Zero-shot RL with Expressivity enhancement (BREEZE), an upgraded FB-based framework that simultaneously enhances learning stability, policy extraction capability, and representation learning quality. BREEZE introduces behavioral regularization in zero-shot RL policy learning, transforming policy optimization into a stable in-sample learning paradigm. Additionally, BREEZE extracts the policy using a task-conditioned diffusion model, enabling the generation of high-quality and multimodal action distributions in zero-shot RL settings. Moreover, BREEZE employs expressive attention-based architectures for representation modeling to capture the complex relationships between environmental dynamics. Extensive experiments on ExORL and D4RL Kitchen demonstrate that BREEZE achieves the best or near-the-best performance while exhibiting superior robustness compared to prior offline zero-shot RL methods. The official implementation is available at: https://github.com/Whiterrrrr/BREEZE.

The remarkable progress of Large Language Models (LLMs) presents promising opportunities for Verilog code generation which is significantly important for automated circuit design. The lacking of meaningful functional rewards hinders the preference optimization based on Reinforcement Learning (RL) for producing functionally correct Verilog code. In this paper, we propose Signal-Aware Learning for Verilog code generation (QiMeng-SALV) by leveraging code segments of functionally correct output signal to optimize RL training. Considering Verilog code specifies the structural interconnection of hardware gates and wires so that different output signals are independent, the key insight of QiMeng-SALV is to extract verified signal-aware implementations in partially incorrect modules, so as to enhance the extraction of meaningful functional rewards. Roughly, we verify the functional correctness of signals in generated module by comparing with that of reference module in the training data. Then abstract syntax tree (AST) is employed to identify signal-aware code segments which can provide meaningful functional rewards from erroneous modules. Finally, we introduce signal-aware DPO which is optimized on the correct signal-level code segments, thereby preventing noise and interference from incorrect signals. The proposed QiMeng-SALV underscores the paradigm shift from conventional module-level to fine-grained signal-level optimization in Verilog code generation, addressing the issue of insufficient functional rewards. Experiments demonstrate that our method achieves state-of-the-art performance on VerilogEval and RTLLM, with a 7B parameter model matching the performance of the DeepSeek v3 671B model and significantly outperforming the leading open-source model CodeV trained on the same dataset.

In both neuroscience and artificial intelligence (AI), it is well-established that neural “coupling” gives rise to dynamically distributed systems. These systems exhibit self-organized spatiotemporal patterns of synchronized neural oscillations, enabling the representation of abstract concepts. By capitalizing on the unprecedented amount of human neuroimaging data, we propose that advancing the theoretical understanding of rhythmic coordination in neural circuits can offer powerful design principles for the next generation of machine learning models with improved efficiency and robustness. To this end, we introduce a physics-informed deep learning framework for \underline{B}rain \underline{R}hythm \underline{I}dentification by \underline{K}uramoto and \underline{C}ontrol (coined \modelname{}) to characterize the synchronization of neural oscillations that shapes the dynamics of evolving cognitive states. Recognizing that brain networks are structurally connected yet behaviorally dynamic, we further conceptualize rhythmic neural activity as an artificial dynamical system of coupled oscillators, offering a shared mechanistic bridge to brain-inspired machine intelligence. By treating each node as an oscillator interacting with its neighbors, this approach moves beyond the conventional paradigm of graph heat diffusion and establishes a new regime of representation compression through oscillatory synchronization. Empirical evaluations demonstrate that this synchronization-driven mechanism not only mitigates over-smoothing in deep GNNs but also enhances the model’s capacity for reasoning and solving complex graph-based problems.

Recent advances in LLM-based decompilers have been shown effective to convert low-level binaries into human-readable source code. However, there still lacks a comprehensive benchmark that provides large-scale binary-source function pairs, which is critical for advancing the LLM decompilation technology. Creating accurate binary-source mappings incurs severe issues caused by complex compilation settings and widespread function inlining that obscure the correspondence between binaries and their original source code. Previous efforts have either relied on used contest‐style benchmarks, synthetic binary–source mappings that diverge significantly from the mappings in real world, or partially matched binaries with only code lines or variable names, compromising the effectiveness of analyzing the binary functionality. To alleviate these issues, we introduce Decompile-Bench, the first open-source dataset comprising two million binary-source function pairs condensed from 100 million collected function pairs, i.e., 450GB of binaries compiled from permissively licensed GitHub projects. For the evaluation purposes, we also developed a benchmark Decompile-Bench-Eval including manually crafted binaries from the well-established HumanEval and MBPP, alongside the compiled GitHub repositories released after 2025 to mitigate data leakage issues. We further explore commonly-used evaluation metrics to provide a thorough assessment of the studied LLM decompilers and find that fine-tuning with Decompile-Bench causes a 20% improvement over previous benchmarks in terms of the re-executability rate. Our code and data has been released in HuggingFace and Github. https://github.com/anonepo/LLM4Decompile

Meta-learning has been demonstrated to be useful to improve the sampling efficiency of Bayesian optimization (BO) and surrogate-assisted evolutionary algorithms (SAEAs) when solving expensive optimization problems (EOPs). Existing studies mainly focus on either combinations of existing meta-learning modeling methods with optimization algorithms, or the development of meta-learning acquisition functions for specific meta BO. However, the meta-learning models used in the literature are not designed for optimization purpose, and the generalization ability of meta-learning acquisition functions is limited. In this work, we develop a novel architecture of meta-learning model for optimization purpose and propose a generalized few-shot evolutionary optimization (FSEO) framework to solve EOPs. We focus on the scenario of expensive multi-objective EOPs (EMOPs) in the context of few-shot optimization as there are few studies on it and its high requirement on surrogate modeling performance. The surrogates in FSEO framework combines neural network with Gaussian Processes (GPs), their network parameters and some parameters of GPs represent task-independent experience and are meta-learned across related optimization tasks, the remaining GPs parameters are task-specific parameters that represent unique features of the target task. We demonstrate that our FSEO framework is able to improve the sampling efficiency of existing SAEAs on EMOPs.