NeurIPS 2025 Papers

Generalized Additive Models (GAMs) are commonly considered *interpretable* within the ML community, as their structure makes the relationship between inputs and outputs relatively understandable. Therefore, it may seem natural to hypothesize that obtaining meaningful explanations for GAMs could be performed efficiently and would not be computationally infeasible. In this work, we challenge this hypothesis by analyzing the *computational complexity* of generating different explanations for various forms of GAMs across multiple contexts. Our analysis reveals a surprisingly diverse landscape of both positive and negative complexity outcomes. Particularly, under standard complexity assumptions such as P$\neq$NP, we establish several key findings: (1) in stark contrast to many other common ML models, the complexity of generating explanations for GAMs is heavily influenced by the structure of the input space; (2) the complexity of explaining GAMs varies significantly with the types of component models used - but interestingly, these differences only emerge under specific input domain settings; (3) significant complexity distinctions appear for obtaining explanations in regression tasks versus classification tasks in GAMs; and (4) expressing complex models like neural networks additively (e.g., as neural additive models) can make them easier to explain, though interestingly, this benefit appears only for certain explanation methods and input domains. Collectively, these results shed light on the feasibility of computing diverse explanations for GAMs, offering a rigorous theoretical picture of the conditions under which such computations are possible or provably hard.

Prevailing scene coordinate regression methods for LiDAR localization suffer from localization ambiguities, as distinct locations can exhibit similar geometric signatures — a challenge that current geometry-based regression approaches have yet to solve. Recent vision–language models show that textual descriptions can enrich scene understanding, supplying potential localization cues missing from point cloud geometries. In this paper, we propose GTR-Loc, a novel text-assisted LiDAR localization framework that effectively generates and integrates geospatial text regularization to enhance localization accuracy. We propose two novel designs: a Geospatial Text Generator that produces discrete pose-aware text descriptions, and a LiDAR-Anchored Text Embedding Refinement module that dynamically constructs view-specific embeddings conditioned on current LiDAR features. The geospatial text embeddings act as regularization to effectively reduce localization ambiguities. Furthermore, we introduce a Modality Reduction Distillation strategy to transfer textual knowledge. It enables high-performance LiDAR-only localization during inference, without requiring runtime text generation. Extensive experiments on challenging large-scale outdoor datasets, including QEOxford, Oxford Radar RobotCar, and NCLT, demonstrate the effectiveness of GTR-Loc. Our method significantly outperforms state-of-the-art approaches, notably achieving a 9.64%/8.04% improvement in position/orientation accuracy on QEOxford. Our code is available at https://github.com/PSYZ1234/GTR-Loc.

Physics-informed machine learning is gaining significant traction for enhancing statistical performance and sample efficiency through the integration of physical knowledge. However, current theoretical analyses often presume complete prior knowledge in non-hybrid settings, overlooking the crucial integration of observational data, and are frequently limited to linear systems, unlike the prevalent nonlinear nature of many real-world applications. To address these limitations, we introduce a discrete weak form that unifies collocation and variational methods, enabling the incorporation of incomplete and complex physical constraints in hybrid learning settings. Within this formulation, we establish that the generalization performance of physics-informed regression in such hybrid settings is governed by the dimension of the affine variety associated with the physical constraint, rather than by the number of parameters. This enables a unified analysis that is applicable to both linear and nonlinear equations. We also present a method to approximate this dimension and provide experimental validation of our theoretical findings.

In this paper, we introduce BMMR, a large-scale bilingual, multimodal, multi-disciplinary reasoning dataset for the community to develop and evaluate large multimodal models (LMMs). BMMR comprises 100k university-level questions drawn from 300 UNESCO-defined subjects, spanning diverse formats—multiple-choice, fill-in-the-blank, and open-ended QA—and sourced from both print and digital media such as books, exams, and quizzes. All data are curated and filtered via a human-in-the-loop, automated, and scalable framework, and each instance is paired with a high-quality reasoning path. The dataset is organized into two parts: BMMR-Eval that comprises 20k high-quality instances to comprehensively assess LMMs’ knowledge and reasoning across multiple disciplines in both Chinese and English; and BMMR-Train that contains 80k instances to support further research and development, extending the current focus on mathematical reasoning to diverse disciplines and domains. In addition, we propose the process-based multi-discipline BMMR-Verifier for accurate and fine-grained evaluation of LMMs’ reasoning. Extensive experiments reveal that (i) even SOTA models leave substantial headroom on BMMR-Eval; (ii) reasoning models exhibit discipline bias and outperform LMMs only on specific subjects; (iii) open-source models still trail their proprietary counterparts; and (iv) fine-tuning on BMMR-Train narrows this gap. Additionally, we conduct reasoning-chain analyses using BMMR-Verifier and other in-depth studies, uncovering the challenges LMMs currently face in multidisciplinary reasoning. We will release the data and models, and we believe our work can offers valuable insights and contributions to the community.

Deductive formal problem-solving (D-FPS) enables process-verified, human-aligned problem-solving by implementing deductive solving processes within formal theorem proving (FTP) environments. However, current methods fail to address the misalignment between informal and formal reasoning granularity and suffer from inefficiency due to backtracking and error propagation. Moreover, the extreme scarcity of formal problem-solution pairs further hinders progress. For the first gap, we propose **HAR** (_**H**ierarchical **A**utoregressive Formal **R**easoner_), a novel reasoning pipeline. HAR decouples informal-aligned drafting and detailed proving, and formulates solution construction as autoregressive generation with per-step feedback. Second, we propose **CoPA** (_**C**hain-**o**f-**P**roxy-**A**utoformalization_), a data generation pipeline that cascades statement autoformalization, proof drafting, and proof search as a proxy autoformalization path. Experiments demonstrate significant improvements: trained on data bootstrapped by CoPA, HAR achieves superior performance on FormalMath500 ($15.50\\%\mapsto 44.09\\%$) and MiniF2F-Solving ($21.87\\%\mapsto 56.58\\%$) with lower computational budget. Explorations reveal promising directions in formal solution pruning and informal dataset denoising.

Identifying and measuring biases associated with sensitive attributes is a crucial consideration in healthcare to prevent treatment disparities. One prominent issue is inaccurate pulse oximeter readings, which tend to overestimate oxygen saturation for dark-skinned patients and misrepresent supplemental oxygen needs. Most existing research has revealed *statistical disparities* linking device measurement errors to patient outcomes in intensive care units (ICUs) without causal formalization. This study *causally* investigates how racial discrepancies in oximetry measurements affect invasive ventilation in ICU settings. We employ a causal inference-based approach using *path-specific effects* to isolate the impact of bias by race on clinical decision-making. To estimate these effects, we leverage a doubly robust estimator, propose its self-normalized variant for improved sample efficiency, and provide novel finite-sample guarantees. Our methodology is validated on semi-synthetic data and applied to two large real-world health datasets: MIMIC-IV and eICU. Contrary to prior work, our analysis reveals minimal impact of racial discrepancies on invasive ventilation rates. However, path-specific effects mediated by oxygen saturation disparity are more pronounced on ventilation duration, and the severity differs by dataset. Our work provides a novel pipeline for investigating potential disparities in clinical decision-making and, more importantly, highlights the necessity of causal methods to robustly assess fairness in healthcare.

Envision an AI capable of functioning in human-like settings, moving beyond mere observation to actively understand, anticipate, and proactively respond to unfolding events. Towards this vision, we focus on the innovative task where, given ego-streaming video input, an assistant proactively answers diverse, evolving questions at the opportune moment, while maintaining synchronized perception and reasoning. This task embodies three key properties: (1) Proactive Coherence, (2) Just-in-Time Responsiveness, and (3) Synchronized Efficiency. To evaluate and address these properties, we first introduce ESTP-Bench (Ego Streaming Proactive Benchmark) alongside the ESTP-F1 metric—a novel framework designed for their rigorous assessment. Secondly, we propose a comprehensive technical pipeline to enable models to tackle this challenging task. This pipeline comprises: (1) a data engine, (2) a multi-stage training strategy, and (3) a proactive dynamic compression technique. Our proposed model effectively addresses these critical properties while achieving state-of-the-art (SOTA) performance on the standard COIN benchmark.

Large language models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their ability to generate human-like text has raised concerns about potential misuse. This underscores the need for reliable and effective methods to detect LLM-generated text. In this paper, we propose IRM, a novel zero-shot approach that leverages Implicit Reward Models for LLM-generated text detection. Such implicit reward models can be derived from publicly available instruction-tuned and base models. Previous reward-based method relies on preference construction and task-specific fine-tuning. In comparison, IRM requires neither preference collection nor additional training. We evaluate IRM on the DetectRL benchmark and demonstrate that IRM can achieve superior detection performance, outperforms existing zero-shot and supervised methods in LLM-generated text detection.

Modern machine learning often requires training with large batch size, distributed data, and massively parallel compute hardware (like mobile and other edge devices or distributed data centers). Communication becomes a major bottleneck in such settings but methods like Local Stochastic Gradient Descent (Local SGD) show great promise to reduce the global communication need. Local SGD consists of three parts: a local optimization processes, an aggregation mechanism, and an outer optimizer that uses the aggregated updates from the nodes to produce a new model. While there exists an extensive literature on understanding the impact of hyperparameters in the local optimization process, the choice of outer optimizer and its hyperparameters is less clear. We study the role of the outer learning in Local SGD, and prove new convergence guarantees for the algorithm. In particular, we show that tuning the outer learning rate allows us to (a) trade off between optimization error and stochastic gradient noise variance, and (b) make up for ill-tuning of the inner learning rate. Our theory suggests that the outer learning rate should sometimes be set to values greater than $1$. We extend our results to apply to when we use momentum in the outer optimizer, and also introduce a novel data-dependent analysis of Local SGD that yields further insights on outer learning rate tuning. We conduct comprehensive experiments with standard language models and various outer optimizers to validate our theory.

Current text-to-image (T2I) models often fail to account for diverse human experiences, leading to misaligned systems. We advocate for pluralism in AI alignment, where an AI understands and is steerable towards diverse, and often conflicting, human values. Our work provides three core contributions to achieve this in T2I models. First, we introduce a novel dataset for Diverse Intersectional Visual Evaluation (DIVE) -- the first multimodal dataset for pluralistic alignment. It enables deep alignment to diverse safety perspectives through a large pool of demographically intersectional human raters who provided extensive feedback across 1000 prompts, with high replication, capturing nuanced safety perceptions. Second, we empirically confirm demographics as a crucial proxy for diverse viewpoints in this domain, revealing significant, context-dependent differences in harm perception that diverge from conventional evaluations. Finally, we discuss implications for building aligned T2I models, including efficient data collection strategies, LLM judgment capabilities, and model steerability towards diverse perspectives. This research offers foundational tools for more equitable and aligned T2I systems. Content Warning: The paper includes sensitive content that may be harmful.

Scalable Vector Graphics (SVG) offer a powerful format for representing visual designs as interpretable code. Recent advances in vision-language models (VLMs) have enabled high-quality SVG generation by framing the problem as a code generation task and leveraging large-scale pretraining. VLMs are particularly suitable for this task as they capture both global semantics and fine-grained visual patterns, while transferring knowledge across vision, natural language, and code domains. However, existing VLM approaches often struggle to produce faithful and efficient SVGs because they never observe the rendered images during training. Although differentiable rendering for autoregressive SVG code generation remains unavailable, rendered outputs can still be compared to original inputs, enabling evaluative feedback suitable for reinforcement learning (RL). We introduce Reinforcement Learning from Rendering Feedback, an RL method that enhances SVG generation in autoregressive VLMs by leveraging feedback from rendered SVG outputs. Given an input image, the model generates SVG roll-outs that are rendered and compared to the original image to compute a reward. This visual fidelity feedback guides the model toward producing more accurate, efficient, and semantically coherent SVGs. \method significantly outperforms supervised fine-tuning, addressing common failure modes and enabling precise, high-quality SVG generation with strong structural understanding and generalization.

Recent interpretability work on large language models (LLMs) has been increasingly dominated by a feature-discovery approach with the help of proxy modules. Then, the quality of features learned by, e.g., sparse auto-encoders (SAEs), is evaluated. This paradigm naturally raises a critical question: do such learned features have better properties than those already represented within the original model parameters, and unfortunately, only a few studies have made such comparisons systematically so far. In this work, we revisit the interpretability of feature vectors stored in feed-forward (FF) layers, given the perspective of FF as key-value memories, with modern interpretability benchmarks. Our extensive evaluation revealed that SAE and FFs exhibits a similar range of interpretability, although SAEs displayed an observable but minimal improvement in some aspects. Furthermore, in certain aspects, surprisingly, even vanilla FFs yielded better interpretability than the SAEs, and features discovered in SAEs and FFs diverged. These bring questions about the advantage of SAEs from both perspectives of feature quality and faithfulness, compared to directly interpreting FF feature vectors, and FF key-value parameters serve as a strong baseline in modern interpretability research.

Vision-language foundation models (VLMs), such as CLIP, exhibit remarkable performance across a wide range of tasks. However, deploying these models can be unreliable when significant distribution gaps exist between training and test data, while fine-tuning for diverse scenarios is often costly. Cache-based test-time adapters offer an efficient alternative by storing representative test samples to guide subsequent classifications. Yet, these methods typically employ naive cache management with limited capacity, leading to severe catastrophic forgetting when samples are inevitably dropped during updates. In this paper, we propose DOTA (DistributiOnal Test-time Adaptation), a simple yet effective method addressing this limitation. Crucially, instead of merely memorizing individual test samples, DOTA continuously estimates the underlying distribution of the test data stream. Test-time posterior probabilities are then computed using these dynamically estimated distributions via Bayes' theorem for adaptation. This distribution-centric approach enables the model to continually learn and adapt to the deployment environment. Extensive experiments validate that DOTA significantly mitigates forgetting and achieves state-of-the-art performance compared to existing methods.

Length generalization—the ability to solve problems longer than those seen during training—remains a critical challenge for large language models (LLMs). Previous work modifies positional encodings (PEs) and data formats to improve length generalization on specific symbolic tasks such as addition and sorting. However, these approaches are fundamentally limited to special tasks, often degrading general language performance. Furthermore, they are typically evaluated on small transformers trained from scratch on single tasks and can cause performance drop when applied during post-training stage of practical LLMs with general capabilities. Hu et al., (2024) proposed Rule-Following Fine-Tuning (RFFT) to improve length generalization in the post-training stage of LLMs. Despite its compatibility with practical models and strong performance, RFFT is proposed for single tasks too, requiring re-training for each individual task with extensive examples. In this paper, we study length generalization in multi-task settings and propose *Meta Rule-Following Fine-Tuning (Meta-RFFT)*, the first framework enabling robust *cross-task* length generalization. As our first contribution, we construct a large length generalization dataset containing **86 tasks** spanning code execution, number processing, symbolic and logical reasoning tasks, beyond the common addition or multiplication tasks. Secondly, we show that cross-task length generalization is possible with Meta-RFFT—after training on a large number of tasks and instances, the models achieve remarkable length generalization ability on *unseen* tasks with *minimal fine-tuning or one-shot prompting*. For example, after fine-tuning on 1 to 5 digit addition, our 32B model **achieves 95% accuracy on 30 digit addition**, significantly outperforming the state-of-the-art reasoning models (DeepSeek-R1-671B: 72%; QwQ-32B: 32%), despite never seeing this task during RF-pretraining.

We study the problem of controlling worst-case errors in misspecified linear regression under the random design setting, where the regression function is estimated via (penalized) least-squares. This setting arises naturally in value function approximation for bandit algorithms and reinforcement learning (RL). Our first main contribution is the observation that the amplification of the misspecification error when using least-squares is governed by the \emph{Lebesgue constant}, a classical quantity from approximation theory that depends on the choice of the feature subspace and the covariate distribution. We also show that this dependence on the misspecification error is tight for least-squares regression: in general, no method minimizing the empirical squared loss, including regularized least-squares, can improve it substantially. We argue this explains the empirical observation that some feature-maps (e.g., those derived from the Fourier bases) ``work better in RL'' than others (e.g., polynomials): given some covariate distribution, the Lebesgue constant is known to be highly sensitive to choice of the feature-map. As a second contribution, we propose a method that augments the original feature set with auxiliary features designed to reduce the error amplification. We then prove that the method successfully competes with an "oracle'' that knows the best way of using the auxiliary features to reduce this amplification. For example, when the domain is a real interval and the features are monomials, our method reduces the amplification factor to $O(1)$ as $d\to\infty$, while without our method, least-squares with the monomials (and in fact polynomials) will suffer a worst-case error amplification of order $\Omega(d)$. It follows that there are functions and feature maps for which our method is consistent, while least-squares is inconsistent.

Programming assistants powered by large language models have transformed software development, yet most benchmarks focus narrowly on code generation tasks. Recent efforts like InfiBench and StackEval attempt to address this gap using Stack Overflow data but remain limited to single-turn interactions in isolated contexts, require significant manual curation, and fail to represent complete project environments. We introduce CodeAssistBench (CAB), the first benchmark framework for evaluating multi-turn programming assistance in realistic settings that address questions grounded in actual codebases. Unlike existing programming Q&A benchmarks, CAB automatically generates scalable datasets from GitHub issues tagged with questions using configurable parameters (e.g., repository creation date, star count, programming languages), and includes automatic containerization of codebases for evaluation. It then evaluates models through simulated users in these containerized environments with full codebase access. Using this framework, we constructed a test set of 3,286 real-world programming questions across 214 repositories, spanning seven programming languages and diverse problem domains. Our evaluation of leading LLMs reveals a substantial capability gap: while models perform well on Stack Overflow questions with success rates of 70-83%, they resolve only up to 16.49% of CAB's issues from recent repositories (post-training cutoff). This discrepancy highlights the challenges of providing assistance in complex, project-specific contexts versus answering standalone questions. Our fully automated framework enables continuous benchmark expansion and is available at https://github.com/amazon-science/CodeAssistBench/.

Recent advancements in Large Language Models (LLMs) have emphasized the critical role of fine-tuning (FT) techniques in adapting LLMs to specific tasks, especially when retraining from scratch is computationally infeasible. Fine-tuning enables LLMs to leverage task- or domain-specific data, producing models that more effectively meet the requirements of targeted applications. However, con- ventional FT approaches often suffer from catastrophic forgetting and suboptimal data efficiency, limiting their real-world applicability. To address these challenges, this paper proposes DEAL, a novel framework that integrates Low-Rank Adapta- tion (LoRA) with a continuous fine-tuning strategy. By incorporating knowledge retention and adaptive parameter update modules, the framework mitigates the limitations of existing FT methods while maintaining efficiency. Experiments on 15 diverse datasets show that DEAL consistently outperforms baseline methods, yielding substantial gains in task accuracy and resource efficiency. These find- ings demonstrate the potential of our approach to advance continual adaptation in LLMs by enhancing task performance while improving resource efficiency. The source code is publicly available at https://github.com/Applied-Machine-Learning-Lab/DEAL.

Series Section Electron Microscopy (ssEM) has emerged as a pivotal technology for deciphering nanoscale biological architectures. Three-dimensional (3D) registration is a critical step in ssEM, tasked with rectifying axial misalignments and nonlinear distortions introduced during serial sectioning. The core scientific challenge lies in achieving distortion mitigation without erasing the natural morphological deformations of biological tissues, thereby enabling faithful reconstruction of 3D ultrastructural organization. In this study, we present a paradigm-shifting optimization framework that rethinks 3D registration through the lens of manifold trajectory optimization. We propose the first continuous trajectory dynamics formulation for 3D registration and introduce a novel optimization strategy. Specifically, we introduce a dual optimization objective that inherently balances global trajectory smoothness with local structural preservation, while developing a solver that combines Gauss-Seidel iteration with Neural ODEs to systematically integrate biophysical priors with data-driven deformation compensation. A key strength of our method is its fully unsupervised training, which avoids reliance on ground truth and suits ssEM scenarios where annotations are difficult to obtain. Extensive experiments on multiple datasets spanning diverse tissue types demonstrate our method's superior performance in structural restoration accuracy and cross-tissue robustness.

The advancement of Embodied AI heavily relies on large-scale, simulatable 3D scene datasets characterized by scene diversity and realistic layouts. However, existing datasets typically suffer from limitations in data scale or diversity, sanitized layouts lacking small items, and severe object collisions. To address these shortcomings, we introduce \textbf{InternScenes}, a novel large-scale simulatable indoor scene dataset comprising approximately 40,000 diverse scenes by integrating three disparate scene sources, \ie, real-world scans, procedurally generated scenes, and designer-created scenes, including 1.96M 3D objects and covering 15 common scene types and 288 object classes. We particularly preserve massive small items in the scenes, resulting in realistic and complex layouts with an average of 41.5 objects per region. Our comprehensive data processing pipeline ensures simulatability by creating real-to-sim replicas for real-world scans, enhances interactivity by incorporating interactive objects into these scenes, and resolves object collisions by physical simulations. We demonstrate the value of InternScenes with two benchmark applications: scene layout generation and point-goal navigation. Both show the new challenges posed by the complex and realistic layouts. More importantly, InternScenes paves the way for scaling up the model training for both tasks, making the generation and navigation in such complex scenes possible. We commit to open-sourcing the data and benchmarks to benefit the whole community.