NeurIPS 2025 Papers

Out-of-town trip recommendation aims to generate a sequence of Points of Interest (POIs) for users traveling from their hometowns to previously unvisited regions based on personalized itineraries, e.g., origin, destination, and trip duration. Modeling the complex user preferences--which often exhibit a two-fold nature of static and dynamic interests--is critical for effective recommendations. However, the sparsity of out-of-town check-in data presents significant challenges in capturing such user preferences. Meanwhile, existing methods often conflate the static and dynamic preferences, resulting in suboptimal performance. In this paper, we for the first time systematically study the problem of out-of-town trip recommendation. A novel framework SPOT-Trip is proposed to explicitly learns the dual static-dynamic user preferences. Specifically, to handle scarce data, we construct a POI attribute knowledge graph to enrich the semantic modeling of users’ hometown and out-of-town check-ins, enabling the static preference modeling through attribute relation-aware aggregation. Then, we employ neural ordinary differential equations (ODEs) to capture the continuous evolution of latent dynamic user preferences and innovatively combine a temporal point process to describe the instantaneous probability of each preference behavior. Further, a static-dynamic fusion module is proposed to merge the learned static and dynamic user preferences. Extensive experiments on real data offer insight into the effectiveness of the proposed solutions, showing that SPOT-Trip achieves performance improvement by up to 17.01%.

In recommendation systems, implicit feedback data can be automatically recorded and is more common than explicit feedback data. However, implicit feedback poses two challenges for relevance prediction, namely (a) positive-unlabeled (PU): negative feedback does not necessarily imply low relevance and (b) missing not at random (MNAR): items that are popular or frequently recommended tend to receive more clicks than other items, even if the user does not have a significant interest in them. Existing methods either overlook the MNAR issue or fail to account for the inherent mechanism of the PU issue. As a result, they may lead to inaccurate relevance predictions or inflated biases and variances. In this paper, we formulate the implicit feedback problem as a counterfactual estimation problem with missing treatment variables. Prediction of the relevance in implicit feedback is equivalent to answering the counterfactual question that ``whether a user would click a specific item if exposed to it?". To solve the counterfactual question, we propose the Counterfactual Implicit Feedback (Counter-IF) prediction approach that divides the user-item pairs into four disjoint groups, namely definitely positive (DP), highly exposed (HE), highly unexposed (HU), and unknown (UN) groups. Specifically, Counter-IF first performs missing treatment imputation with different confidence levels from raw implicit feedback, then estimates the counterfactual outcomes via causal representation learning that combines pointwise loss and pairwise loss based on the user-item pairs stratification. Theoretically the generalization bound of the learned model is derived. Extensive experiments are conducted on publicly available datasets to demonstrate the effectiveness of our approach. The code is available at https://github.com/zhouchuanCN/NeurIPS25-Counter-IF.

The Conditional Sequence Modeling (CSM) paradigm, benefiting from the transformer's powerful distribution modeling capabilities, has demonstrated considerable promise in Reinforcement Learning (RL) tasks. However, much of the work has focused on applying CSM to single online or offline settings, with the general architecture rarely explored. Additionally, existing methods primarily focus on deterministic trajectory modeling, overlooking the randomness of state transitions and the diversity of future trajectory distributions. Fortunately, value-based methods offer a viable solution for CSM, further bridging the potential gap between offline and online RL. In this paper, we propose Value-Guided Decision Transformer (VDT), which leverages value functions to perform advantage-weighting and behavior regularization on the Decision Transformer (DT), guiding the policy toward upper-bound optimal decisions during the offline training phase. In the online tuning phase, VDT further integrates value-based policy improvement with behavior cloning under the CSM architecture through limited interaction and data collection, achieving performance improvement within minimal timesteps. The predictive capability of value functions for future returns is also incorporated into the sampling process. Our method achieves competitive performance on various standard RL benchmarks, providing a feasible solution for developing CSM architectures in general scenarios. Code is available at here.

Scientific machine learning often involves representing complex solution fields that exhibit high-frequency features such as sharp transitions, fine-scale oscillations, and localized structures. While implicit neural representations (INRs) have shown promise for continuous function modeling, capturing such high-frequency behavior remains a challenge—especially when modeling multiple solution fields with a shared network. Prior work addressing spectral bias in INRs has primarily focused on single-instance settings, limiting scalability and generalization. In this work, we propose Global Fourier Modulation (GFM), a novel modulation technique that injects high-frequency information at each layer of the INR through Fourier-based reparameterization. This enables compact and accurate representation of multiple solution fields using low-dimensional latent vectors. Building upon GFM, we introduce PDEfuncta, a meta-learning framework designed to learn multi-modal solution fields and support generalization to new tasks. Through empirical studies on diverse scientific problems, we demonstrate that our method not only improves representational quality but also shows potential for forward and inverse inference tasks without the need for retraining.

Text-to-image (T2I) models raise ethical and safety concerns due to their potential to generate inappropriate or harmful images. Evaluating these models' security through red-teaming is vital, yet white-box approaches are limited by their need for internal access, complicating their use with closed-source models. Moreover, existing black-box methods often assume knowledge about the model's specific defense mechanisms, limiting their utility in real-world commercial API scenarios. A significant challenge is how to evade unknown and diverse defense mechanisms. To overcome this difficulty, we propose a novel Rule-based Preference modeling Guided Red-Teaming (RPG-RT), which iteratively employs LLM to modify prompts to query and leverages feedback from T2I systems for fine-tuning the LLM. RPG-RT treats the feedback from each iteration as a prior, enabling the LLM to dynamically adapt to unknown defense mechanisms. Given that the feedback is often labeled and coarse-grained, making it difficult to utilize directly, we further propose rule-based preference modeling, which employs a set of rules to evaluate desired or undesired feedback, facilitating finer-grained control over the LLM’s dynamic adaptation process. Extensive experiments on nineteen T2I systems with varied safety mechanisms, three online commercial API services, and T2V models verify the superiority and practicality of our approach. Our codes are available at: https://github.com/caosip/RPG-RT.

Modeling interactive driving behaviors in complex scenarios remains a fundamental challenge for autonomous driving planning. Learning-based approaches attempt to address this challenge with advanced generative models, removing the dependency on over-engineered architectures for representation fusion. However, brute-force implementation by simply stacking transformer blocks lacks a dedicated mechanism for modeling interactive behaviors that is common in real driving scenarios. The scarcity of interactive driving data further exacerbates this problem, leaving conventional imitation learning methods ill-equipped to capture high-value interactive behaviors. We propose Flow Planner, which tackles these problems through coordinated innovations in data modeling, model architecture, and learning scheme. Specifically, we first introduce fine-grained trajectory tokenization, which decomposes the trajectory into overlapping segments to decrease the complexity of whole trajectory modeling. With a sophisticatedly designed architecture, we achieve efficient temporal and spatial fusion of planning and scene information, to better capture interactive behaviors. In addition, the framework incorporates flow matching with classifier-free guidance for multi-modal behavior generation, which dynamically reweights agent interactions during inference to maintain coherent response strategies, providing a critical boost for interactive scenario understanding. Experimental results on the large-scale nuPlan dataset demonstrate that Flow Planner achieves state-of-the-art performance among learning-based approaches while effectively modeling interactive behaviors in complex driving scenarios.

We present EyeBench, the first benchmark designed to evaluate machine learning models that decode cognitive and linguistic information from eye movements during reading. EyeBench offers an accessible entry point to the challenging and underexplored domain of modeling eye tracking data paired with text, aiming to foster innovation at the intersection of multimodal AI and cognitive science. The benchmark provides a standardized evaluation framework for predictive models, covering a diverse set of datasets and tasks, ranging from assessment of reading comprehension to detection of developmental dyslexia. Progress on the EyeBench challenge will pave the way for both practical real-world applications, such as adaptive user interfaces and personalized education, and scientific advances in understanding human language processing. The benchmark is released as an open-source software package which includes data downloading and harmonization scripts, baselines and state-of-the-art models, as well as evaluation code, publicly available at https://github.com/EyeBench/eyebench.

Crystal structures are defined by the periodic arrangement of atoms in 3D space, inherently making them equivariant to SO(3) group. A fundamental requirement for crystal property prediction is that the model's output should remain invariant to arbitrary rotational transformations of the input structure. One promising strategy to achieve this invariance is to align the given crystal structure into a canonical orientation with appropriately computed rotations, or called frames. However, existing work either only considers a global frame or solely relies on more advanced local frames based on atoms' local structure. A global frame is too coarse to capture the local structure heterogeneity of the crystal, while local frames may inadvertently disrupt crystal symmetry, limiting their expressivity. In this work, we revisit the frame design problem for crystalline materials and propose a novel approach to construct expressive {\bf S}ymmetry Preserving Frames, dubbed as SPFrame, for modeling crystal structures. Specifically, this local-global associative frame constructs invariant local frames rather than equivariant ones, thereby preserving the symmetry of the crystal. In parallel, it integrates global structural information to construct an equivariant global frame to enforce SO(3) invariance. Extensive experimental results demonstrate that SPFrame consistently outperforms traditional frame construction techniques and existing crystal property prediction baselines across multiple benchmark tasks.

Tokenization imposes a fixed granularity on the input text, freezing how a language model operates on data and how far in the future it predicts. Byte Pair Encoding (BPE) and similar schemes split text once, build a static vocabulary, and leave the model stuck with that choice. We relax this rigidity by introducing an autoregressive U-Net that learns to embed its own tokens as it trains. The network reads raw bytes, pools them into words, then pairs of words, then up to 4 words, giving it a multi-scale view of the sequence. At deeper stages, the model must predict further into the future -- anticipating the next few words rather than the next byte -- so deeper stages focus on broader semantic patterns while earlier stages handle fine details. When carefully tuning and controlling pretraining compute, shallow hierarchies tie strong BPE baselines, and deeper hierarchies have a promising trend. Because tokenization now lives inside the model, the same system can handle character-level tasks and carry knowledge across low-resource languages.

Chemical reaction prediction remains a fundamental challenge in organic chemistry, where existing machine learning models face two critical limitations: sensitivity to input permutations (molecule/atom orderings) and inadequate modeling of substructural interactions governing reactivity. These shortcomings lead to inconsistent predictions and poor generalization to real-world scenarios. To address these challenges, we propose ReaDISH, a novel reaction prediction model that learns permutation-invariant representations while incorporating interaction-aware features. It introduces two innovations: (1) symmetric difference shingle encoding, which computes molecular shingle differences to capture reaction-specific structural changes while eliminating order sensitivity; and (2) geometry-structure interaction attention, a mechanism that models intra- and inter-molecular interactions at the shingle level. Extensive experiments demonstrate that ReaDISH improves reaction prediction performance across diverse benchmarks. It shows enhanced robustness with an average improvement of 8.76\% on R$^2$ under permutation perturbations.

We study how to fine-tune LLMs using user-edit deployment data consisting of a set of context, an agent's response, and user edits. This deployment data is naturally generated by users in applications such as LLMs-based writing assistants and coding agents. The _natural_ origin of user edits makes it a desired source for adapting and personalizing of LLMs. In this setup, there emerges a unification of various feedback types namely preferences, supervised labels, and cost that are typically studied separately in the literature. In this paper, we initiate the theoretical investigation of learning from user edits. We first derive bounds for learning algorithms that learn from each of these feedback types. We prove that these algorithms have different trade-offs depending upon the user, data distribution, and model class. We then propose a simple ensembling procedure to jointly learn from these feedback types. On two domains from Gao et al. 2024, we show our ensembling procedure outperforms these methods that learn from individual feedback. Further, we show that our proposed procedure can robustly adapt to different user-edit distributions at test time.

Advances in large-scale recording technologies now enable simultaneous measurements from multiple brain areas, offering new opportunities to study signal transmission across interacting components of neural circuits. However, neural responses exhibit substantial trial-to-trial variability, often driven by unobserved factors such as subtle changes in animal behavior or internal states. To prevent evolving background dynamics from contaminating identification of functional coupling, we developed a hybrid neural spike train model, GLM-Transformer, that incorporates flexible, deep latent variable models into a point process generalized linear model (GLM) having an interpretable component for cross-population interactions. A Transformer-based variational autoencoder captures nonstationary individual-neuron dynamics that vary across trials, while standard nonparametric regression GLM coupling terms provide estimates of directed interactions between neural populations. We incorporate a low-rank structure on population-to-population coupling effects to improve scalability. Across synthetic datasets and mechanistic simulations, GLM-Transformer recovers known coupling structure and remains robust to shared background fluctuations. When applied to the Allen Institute Visual Coding dataset, it identifies feedforward pathways consistent with established visual hierarchies. This work offers a step toward improved identification of neural population interactions, and contributes to ongoing efforts aimed at achieving interpretable results while harvesting the benefits of deep learning.

Recent breakthroughs in the analysis of multi-objective evolutionary algorithms (MOEAs) are mathematical runtime analyses of those algorithms which are intensively used in practice. So far, most of these results show the same performance as previously known for simple algorithms like the GSEMO. The few results indicating advantages of the popular MOEAs share the same shortages: They consider the performance for the problem of computing the full Pareto front, (of some algorithms enriched with newly invented mechanisms,) and this on newly designed benchmarks. In this work, we overcome these shortcomings by analyzing how existing popular MOEAs approximate the Pareto front of the established LargeFront benchmark. We prove that all popular MOEAs, including NSGA-II (sequential version), NSGA-III, SMS-EMOA, and SPEA2, only need an expected time of $O(n^2 \log n)$ function evaluations to compute an additive $\varepsilon$-approximation of the Pareto front of the LargeFront benchmark. This contrasts with the already proven exponential runtime (with high probability) of the GSEMO on the same task. This result is the first mathematical runtime analysis showing and explaining the superiority of popular MOEAs over simple ones like the GSEMO for the central task of computing good approximations to the Pareto front.

Signed networks can reflect more complex connections through positive and negative edges, and cost-effective signed network sketching can significantly benefit an important link sign prediction task in the era of big data. Existing signed network embedding algorithms mainly learn node representation in the Graph Neural Network (GNN) framework with the balance theory. However, the node-wise representation learning methods either limit the representational power because they primarily rely on node pairwise relationship in the network, or suffer from severe efficiency issues. Recent research has explored simplicial complexes to capture higher-order interactions and integrated them into GNN frameworks. Motivated by that, we propose EdgeSketch+, a simple and effective edge embedding algorithm beyond traditional node-centric modeling that directly represents edges as low-dimensional vectors without transitioning from node embeddings. The proposed approach maintains a good balance between accuracy and efficiency by exploiting the Locality Sensitive Hashing (LSH) technique to swiftly capture the higher-order information derived from the simplicial complex in a manner of no learning processes. Experiments show that EdgeSketch+ matches state-of-the-art accuracy while significantly reducing runtime, achieving speedups of up to $546.07\times$ compared to GNN-based methods.

We propose LiteReality, a novel pipeline that converts RGB-D scans of indoor environments into compact, realistic, and interactive 3D virtual replicas. LiteReality not only reconstructs scenes that visually resemble reality but also supports key features essential for graphics pipelines, such as object individuality, articulation, and high-quality physically based rendering (PBR) materials. At its core, LiteReality first performs scene understanding and parses the results into a coherent 3D layout and object set with the help of a structured scene graph. It then reconstructs the scene by retrieving the most visually similar artist-crafted 3D models from a curated asset database. The Material Painting module further enhances realism by recovering high-quality, spatially varying materials from observed images. Finally, the reconstructed scene is integrated into a simulation engine, where basic physical properties are assigned to enable interactive behavior. The resulting scenes are compact, editable, and fully compatible with standard graphics pipelines, making them suitable for applications in AR/VR, gaming, robotics, and digital-twin systems. In addition, LiteReality introduces a training-free object-retrieval module that achieves state-of-the-art similarity performance on the Scan2CAD dataset, along with a robust Material Painting module capable of transferring appearances from images of any style to 3D assets—even under severe misalignment, occlusion, and poor lighting. We demonstrate the effectiveness of LiteReality on both real-world scans and standard public datasets.

Large language models (LLMs) have shown that generative pretraining can distill vast world knowledge into compact token representations. While LLMs encapsulate extensive world knowledge, they remain limited in modeling the behavioral knowledge contained within user interaction histories. User behavior forms a distinct modality, where each action—defined by multi-dimensional attributes such as time, context, and transaction type—constitutes a behavioral token. Modeling these high-cardinality, sparse, and irregular sequences is challenging, and discriminative models often falter under limited supervision. To bridge this gap, we extend generative pretraining to user behavior, learning transferable representations from unlabeled behavioral data analogous to how LLMs learn from text. We present PANTHER, a hybrid generative–discriminative framework that unifies user behavior pretraining and downstream adaptation, enabling large-scale sequential user representation learning and real-time inference. PANTHER introduces: (1) Structured Tokenization to compress multi-dimensional transaction attributes into an interpretable vocabulary; (2) Sequence Pattern Recognition Module (SPRM) for modeling periodic transaction motifs; (3) a Unified User-Profile Embedding that fuses static demographics with dynamic transaction histories, enabling both personalized predictions and population-level knowledge transfer; and (4) Real-time scalability enabled by offline caching of pre-trained embeddings for millisecond-level inference.Fully deployed and operational online at WeChat Pay, PANTHER delivers a 25.6\% boost in next-transaction prediction HitRate@1 and a 38.6\% relative improvement in fraud detection recall over baselines. Cross-domain evaluations on public benchmarks (CCT, MBD, MovieLens-1M, Yelp) show strong generalization, achieving up to 21\% HitRate@1 gains over transformer baselines, establishing PANTHER as a scalable, high-performance framework for industrial user sequential behavior modeling.

Reinforcement Learning from Human Feedback (RLHF) has shown remarkable success in aligning Large Language Models (LLMs) with human preferences. Traditional RLHF methods rely on a fixed dataset, which often suffers from limited coverage. To this end, online RLHF has emerged as a promising direction, enabling iterative data collection and refinement. Despite its potential, this paradigm faces a key bottleneck: the requirement to continuously integrate new data into the dataset and re-optimize the model from scratch at each iteration, resulting in computational and storage costs that grow linearly with the number of iterations. In this work, we address this challenge by proposing a *one-pass* reward modeling method that eliminates the need to store historical data and achieves constant-time updates per iteration. Specifically, we first formalize RLHF as a contextual preference bandit and develop a new algorithm based on online mirror descent with a tailored local norm, replacing the standard maximum likelihood estimation for reward modeling. We then apply it to various online RLHF settings, including passive data collection, active data collection, and deployment-time adaptation. We provide theoretical guarantees showing that our method enhances both statistical and computational efficiency. Finally, we design practical algorithms for LLMs and conduct experiments with the Llama-3-8B-Instruct and Qwen2.5-7B-Instruct models on Ultrafeedback and Mixture2 datasets, validating the effectiveness of our approach.

Binary choices, as often used for reinforcement learning from human feedback (RLHF), convey only the *direction* of a preference. A person may choose apples over oranges and bananas over grapes, but *which preference is stronger*? Strength is crucial for decision-making under uncertainty and generalization of preference models, but hard to measure reliably. Metadata such as response times and inter-annotator agreement can serve as proxies for strength, but are often noisy and confounded. We propose ResponseRank to address the challenge of learning from noisy strength signals. Our method uses relative differences in these signals to *rank responses to pairwise comparisons by their inferred preference strength*. Signals are only considered locally within carefully constructed strata, controlling for systemic variation. This enables robust learning of utility differences consistent with strength-derived rankings, all while making minimal assumptions. Our contributions are threefold: (1) ResponseRank, a novel method that robustly learns preference strength by leveraging locally valid relative strength signals; (2) empirical evidence of improved sample efficiency and robustness across diverse tasks: synthetic preference learning (with simulated response times), language modeling (with annotator agreement), and RL control tasks (with simulated episode returns); and (3) the *Pearson Distance Correlation (PDC)*, a novel metric that isolates cardinal utility learning from ordinal accuracy.

Recommendation models are predominantly trained using implicit user feedback, since explicit feedback is often costly to obtain. However, implicit feedback, such as clicks, does not always reflect users' real preferences. For example, a user might click on a news article because of its attractive headline, but end up feeling uncomfortable after reading the content. In the absence of explicit feedback, such erroneous implicit signals may severely mislead recommender systems. In this paper, we propose MTRec, a novel sequential recommendation framework designed to align with real user preferences by uncovering their internal satisfaction on recommended items. Specifically, we introduce a mental reward model to quantify user satisfaction and propose a distributional inverse reinforcement learning approach to learn it. The learned mental reward model is then used to guide recommendation models to better align with users’ real preferences. Our experiments show that MTRec brings significant improvements to a variety of recommendation models. We also deploy MTRec on an industrial short video platform and observe a 7\% increase in average user viewing time.