NeurIPS 2026 论文列表

What if you could edit an image simply by drawing on it---circling an object, writing a short label, or sketching an arrow---with minimal or even no typed prompt? We first reveal a surprising capability: when a text-driven image editor is paired with a strong Vision-Language Model (VLM) at inference time, the combined system can directly interpret visual instructions embedded in the image itself, such as on-image text, bounding boxes, and directional arrows. However, this pipeline incurs additional VLM-planning latency, introduces brittle cross-model error propagation, and depends on external APIs or auxiliary model weights. To address these limitations, we introduce Siphon, a framework that internalizes this externally elicited visual-instruction-following competence into a single diffusion-based editor. Rather than relying on an external VLM at inference time, Siphon first uses a VLM planner to synthesize paired visual-instruction supervision, and then transfers this competence into the editor through lightweight LoRA fine-tuning. The resulting model reads, grounds, and executes on-image annotations directly from pixels, while keeping the diffusion sampling cost close to that of the base editor. Across multiple diffusion architectures, Siphon substantially improves spatial controllability, instruction adherence, and marker removal over text-driven baselines, while matching VLM-assisted pipelines without their additional planning latency, API dependence, and cross-model fragility. We do not position rendered long text as a replacement for conventional text prompts; rather, on-image text is mainly intended for short labels and spatially grounded edit intents, while long or complex instructions can still be provided through the standard text channel.

This work identifies a critical failure mode in frontier large language models (LLMs), which we term \textbf{Internal Safety Collapse} (ISC): \textit{under certain task conditions, models enter a state in which they continuously generate large volumes of harmful content while executing otherwise benign tasks}. To systematically study ISC, we introduce \TVD{} (Task, Validator, Data), a framework that instantiates controlled workflows around domain tools, where valid completion requires filling harmful content as structured data. We collect 53 representative workflows across 8 disciplines, from toxicity evaluation to molecular docking and pathogen genome analysis, showing that ISC reproduces across all eight disciplines. In the worst case over three interaction settings, the four frontier LLMs average a \textbf{95.3\%} safety-failure rate. Frontier LLMs with stronger task-completion capability show higher unsafe-completion rates: long-horizon execution skill becomes a liability when workflow completion requires harmful content. We even observe extremely severe harmful content closely resembling outputs from early-generation, unaligned LLMs in 2023. Despite substantial safety alignment efforts, frontier LLMs continue to retain inherently unsafe internal capabilities: alignment reshapes observable outputs but does not eliminate the underlying risk profile. These findings underscore the need for caution when deploying LLMs in high-stakes settings, including scientific pipelines and autonomous agents. Complementary source code is provided at \url{https://anonymous.4open.science/r/NIPS-ISC-Code-Share-BE11}.

Although numerous algorithms have been proposed for unsupervised anomaly detection, there is currently no widely acknowledged internal evaluation method tailored specifically to this task. Existing internal evaluation methods typically assess score distributions alone or depend on auxiliary classifiers, limiting their reliability and interpretability. To address these problems, we propose SSSD (Similar Scores, Similar Data), an evaluation framework inspired by the smoothness assumption in machine learning—that similar inputs should yield similar outputs. We adapt this principle to anomaly detection by requiring that locally similar data points be assigned similar anomaly scores. Instead of testing this assumption directly, SSSD identifies violations of this principle by comparing the similarity of data with similar anomaly scores, interpreting a significant difference as a higher probability of erroneous detection. This evaluation of similarity is measured through nearby-score and nearby-distance, which is calculated based on the distances between neighboring data points and their corresponding anomaly scores. SSSD is highly interpretable, providing valuable insights for identifying mislabeled data. Experimental results show that our proposed method is both effective and robust for evaluating the performance of unsupervised anomaly detection algorithms across artificial and real-world datasets.

Existing research infrastructure is fundamentally document-centric, providing citation links between papers but lacking explicit representations of methodological evolution. In particular, it does not capture the structured relationships that explain how and why research methods emerge, adapt, and build upon one another. With the rise of AI-driven research agents as a new class of consumers of scientific knowledge, this limitation becomes increasingly consequential, as such agents cannot reliably reconstruct method evolution topologies from unstructured text. We introduce Intern-Atlas, a methodological evolution graph that automatically identifies method-level entities, infers lineage relationships among methodologies, and captures the bottlenecks that drive transitions between successive innovations. Built from $1{,}030{,}314$ papers spanning AI conferences, journals, and arXiv preprints, the resulting graph comprises $9{,}410{,}201$ semantically typed edges, forming a queryable causal network of methodological development. To operationalize this structure, we further propose a self-guided temporal tree search algorithm for constructing evolution chains that trace the progression of methods over time. We evaluate the quality of the resulting graph against expert-curated ground-truth evolution chains and observe strong alignment. In addition, we demonstrate that Intern-Atlas enables downstream applications in idea evaluation and automated idea generation. We position methodological evolution graphs as a foundational data layer for the emerging field of automated scientific discovery.

Emerging latent reasoning paradigms allow Large Language Models to “think” in latent embedding spaces, offering a more efficient alternative to explicit Chain-of-Thought. However, they generally face two critical limitations: First, relying exclusively on latent tokens amplifies uncertainty due to their inherently high entropy, thereby degrading solution correctness. Second, the overthinking problem remains prevalent, and existing approaches typically rely on rigid heuristics for stopping, often resulting in suboptimal termination. To address these challenges, we view efficient reasoning as a learnable control problem over both how to think and when to stop. We formulate this within a Reinforcement Learning framework, training lightweight per-step gates on a frozen LLM. This modulation enables the model to (i) adaptively interleave latent soft-token reasoning with explicit token generation, and (ii) predict cumulative exit probabilities for robust early termination. Experiments on diverse reasoning benchmarks across varying model scales and families demonstrate that our method surpasses both standard CoT and existing latent baselines, achieving accuracy gains of ∼3% while reducing token consumption by up to 40%.

Knowledge Graph Question Answering (KGQA) increasingly requires multi-turn interaction with knowledge graphs (KGs) to derive final answers. Such interactive capabilities often rely on proprietary large-scale LLMs, which hinders cost-efficient and local deployment. Reinforcement learning offers a scalable alternative for endowing smaller open-source LLMs with multi-turn interaction ability, as it enables models to improve from self-explored interaction trajectories rather than relying on expensive expert demonstrations. However, KGQA presents a distinctive challenge for RL: unlike tasks where exploration can proceed in a relatively unconstrained textual space, KG reasoning is governed by a sparse graph topology. Free-form policy rollouts frequently generate relation sequences that cannot be executed on the graph, making reward signals sparse and unstable. We propose ICOR, an Interactive Combinatorial Reinforcement Learning framework that grounds policy exploration in the combinatorial structure of KGs. ICOR first retrieves a query-specific set of candidate relations and then restricts policy rollouts to relation-path compositions within this graph-derived space. This design increases the likelihood that sampled trajectories are executable and thus informative for policy optimization. Building on this constrained rollout space, ICOR applies GRPO with a staged feedback mechanism that guides learning from structural validity to answer-consistent reasoning. Experiments on multiple KGQA datasets demonstrate the effectiveness of ICOR.

Recent advances in robotics foundation models have shown promising multi-task generalization, yet their reliance on specialized embodiment-specific data and model fundamentally prevents them from exploiting the far richer physical interaction knowledge present in abundant human manipulation videos. Existing approaches that leverage human videos suffer from misaligned representations that preclude explicit unified interaction modeling over human video and robot demonstration, failing to capture unified physical interaction dynamics essential for cross-domain transfer. We address this by constructing a unified representation that aligns human videos and robot tasks through scene point clouds with hand-to-gripper mapping via spatial tracking and hand pose estimation. Leveraging this representation, we propose a structured graph transformer that explicitly models spatial, semantic, and intentional interaction entities through graph attention mechanisms. The modeled multi-level interaction features then enable interaction-aligned cross-domain learning, transferring manipulation knowledge from human videos to robot tasks via discrete interaction abstraction and hierarchical distribution alignment. Experiments on RLBench, ManiSkill2, and real-world robot demonstrate state-of-the-art performance on diverse benchmarks with fewer demonstrations and effectiveness of our designed modules. Remarkably, our method achieves zero-shot transfer on robot tasks when trained exclusively on human video, providing strong evidence for effective human-to-robot transfer through interaction alignment.

Existing benchmarks for multimodal agentic search evaluate multimodal search and visual browsing, but visual evidence is either confined to the input or treated as an answer endpoint rather than part of an interleaved search trajectory. We introduce \textbf{InterLV-Search}, a benchmark for Interleaved Language-Vision Agentic Search, in which textual and visual evidence is repeatedly used to condition later search. It contains 2,061 examples across three levels: active visual evidence seeking, controlled offline interleaved multimodal search, and open-web interleaved multimodal search. Beyond existing benchmarks, it also includes multimodal multi-branch samples that involve comparison between multiple entities during the evidence search. We construct Level 1 and Level 2 with automated pipelines and Level 3 with a machine-led, human-supervised open-web pipeline. We further provide InterLV-Agent for standardized tool use, trajectory logging, and evaluation. Experiments on proprietary and open-source multimodal agents show that current systems remain far from solving interleaved multimodal search, with the best model below 50% overall accuracy, highlighting challenges in visual evidence seeking, search control, and multimodal evidence integration. We release the benchmark data and evaluation code at \url{https://anonymous.4open.science/r/InterLV-Search-F423}

The deployment of AI agents in multi-agent workflows enables inter-agent influence, whereby agents strategically steer other agents’ behavior in line with a specific goal. Such capabilities pose novel risks, as malicious agents could direct others toward harmful actions. In this work we investigate three such capabilities –- persuasion, deception, and coercion –- across five realistic evaluation environments and a variety of frontier models. We observe significant inter-agent influence capabilities in frontier models. In oversight environments, all tested models shifted policy-violating decisions from rejection to approval through persuasion, deception, and coercion. In peer-to-peer settings, models extracted concessions in scheduling negotiations and redirected a peer's safety research trajectory toward an attacker-preferred direction. While evidence is weakest for inter-agent coercion, several frontier models independently formulated and executed coercive threats, including threats to harm a named human in the environment. These results point to a pressing need for work on risk mitigation to promote beneficial deployments of multi-agent systems.

As intelligent assistants are increasingly deployed in real-world environments, recommendations move from passive preference matching to grounding user intent from underspecified multimodal evidence. In these settings, users often provide only a reference image and a fuzzy natural-language query, leaving crucial visual attributes, personalized preferences, and situational constraints implicit. Existing recommenders either rely on historical interactions enriched with multimodal item features or presuppose fully specified textual requests, which are insufficient to resolve such queries in a single-turn setting. We present \textbf{IntentLens}, a tool-augmented recommendation framework that grounds underspecified multimodal queries into explicit, query-relevant evidence before ranking. IntentLens adopts a \emph{ground-then-rank} design: a shared multimodal LLM orchestrates visual, user-memory, and item-attribute tools to (i) recover latent user intent from the image and language query, and (ii) enrich each candidate with fine-grained, query-conditioned evidence; a lightweight ranker then scores candidates over the grounded representations. To support this new setting, we further construct two benchmarks with underspecified multimodal queries, \textsc{GoogleReview-MMTool} and \textsc{Yelp-MMTool}. Comprehensive experiments show that IntentLens outperforms strong MLLM and retrieval baselines.

Text-to-CAD generation aims to make CAD modeling more accessible by generating executable CAD programs from natural-language descriptions. However, practical CAD workflows require more than executable geometry: generated programs should also expose reusable structure that supports later modification. We introduce intent-oriented text-to-CAD generation, where design intent is expressed as recoverable code-level signals: semantic features, reusable parameters, and lightweight parametric relations. Rather than changing the model architecture or decoding procedure, we investigate how supervision representation granularity affects both CAD generation and zero-shot intent-preserving editing. To enable this study, we construct IntentCAD-100K, a 100K paired text--code dataset built by converting low-level CAD construction traces into executable CadQuery programs. We further construct IntentCAD-Edit-1K to evaluate zero-shot intent-preserving editing from the same representation perspective. Intent2CAD uses a two-stage conservative lifting pipeline: semantic lifting rewrites reliable primitive groups into feature-level operations, while parametric lifting exposes lightweight parametric relations. Under controlled fine-tuning, semantic lifting mainly improves generation reliability and feature recovery, reducing IR from 2.39% to 0.22% and achieving 99.96% API Recovery, while parametric lifting makes reusable relation structure explicit and further improves relation-aware zero-shot editing.

Obtaining meaningfully diverse high-quality samples from Large Language Models (LLMs) for a fixed prompt remains an open challenge. Current methods often only operate at the token-level, paraphrasing the same response. This is problematic as it leads to poor exploration on reasoning problems and to unengaging, repetitive conversational agents. To address this, we propose Intent Factored Generation (IFG), factorising the sampling process into two stages. First, we sample a semantically dense intent that anchors the sample, e.g., a summary or keywords. Second, we sample the final response conditioning on both the original prompt and the intent from the first stage. This factorisation allows the use a higher temperature during the intent step to promote conceptual diversity, and a lower temperature during the final generation to ensure the outputs are coherent. We find that prompting the model to explicitly state its intent for each step of the chain-of-thought before generating the step is beneficial for reasoning tasks. We show that our simple method is highly effective across a diverse set of tasks. We find that this method improves both exploration and final performance on math and code tasks, and combines well with Reinforcement Learning from Verifier Feedback (RLVF). We also show that our method can be used to increase the diversity of instruction-tuned models. Finally, we demonstrate that our method leads to higher quality diverse samples on a language modelling task, on a new dataset that we open-source. IFG is easy to implement, and can be used with off-the-shelf LLMs to improve the performance-diversity tradeoff. Full Implementation at https://anonymous.4open.science/r/IFG-Anon/

Recent Vision-Language-Action (VLA) models have advanced end-to-end autonomous driving by incorporating reasoning for better interpretability and planning quality. However, most existing approaches directly generate the final trajectory without explicitly examining its future consequences, which limits their reliability in complex and dynamic environments. To address this limitation, we propose **IRR-Drive** (**Intend, Reflect, Refine**), an adaptive multimodal reflection framework for autonomous driving. Specifically, to tightly couple high-level reasoning with physical constraints, IRR-Drive first generates a preliminary textual intention and anticipates potential interactions by predicting future semantic bird's-eye view (BEV) representations. This dual-modality (Text + BEV) reflection space explicitly models anticipated scene evolution, enabling the model to rigorously self-correct and refine its initial intent before generating the final trajectory. Furthermore, to balance planning performance and computational efficiency, we construct reflection-oriented training data and design an adaptive reflection reward, enabling the model to adaptively select its reasoning mode according to scene complexity. Instead of using reasoning primarily as an auxiliary interpretation, IRR-Drive directly integrates an adaptive reflection mechanism into the planning framework, enabling grounded, decision-aware trajectory correction that is driven by scene complexity. Our method achieves state-of-the-art performance on the NAVSIM benchmark in both PDMS and EPDMS. Extensive experiments demonstrate the effectiveness of our multimodal reflection framework and validate the efficacy of the proposed adaptive reflection strategy.

Language models are increasingly deployed as co-scientists in the real world, yet their ability to uphold research integrity, particularly under institutional pressures, remains unmeasured. We introduce IntegrityBench, a comprehensive benchmark evaluating three facets (misconduct classification, ethical action reasoning and artifact-grounded decision making), using 36 paired misconduct and ethical control tasks under a 5-level implicit-explicit pressure protocol across 3 domains and 4 research stages. Through a large-scale evaluation across 18 frontier model variants, we find that, under the strongest pressures, frontier models fail roughly one in three integrity-critical decisions. Surprisingly, neither scale nor reasoning can reliably enhance research integrity. Explicit pressures reliably induce compliance with misconduct while implicit contextual reframing more often causes over refusal of legitimate research tasks. Across tasks, models treat surface-level cues as evidence of misconduct without recognizing procedural justifications. Further, models failing to classify research requests accurately perform equally or better on artifact-grounded decision making (mean accuracy of 85.7 when classification fails versus 79.4 when it succeeds), demonstrating that the three facets might be structurally dissociated because accurate ethical actions don't require correct classification to precede them. Frontier models can appear helpful while displaying research integrity failures that create two distinct deployment risks: facilitating research misconduct and diminishing trust in AI-assisted research outputs.

Randomized controlled trials (RCTs) identify treatment effects for an enrolled trial population by assigning treatment independently of potential outcomes, conditional on baseline covariates. These trial-specific effects may not generalize to a broader target population when the target and trial covariate distributions differ. This problem becomes harder when individual-level target population data are unavailable. Pre-trained large language models (LLMs) offer one way to address this data limitation by generating digital twins (DTs) with synthetic covariates intended to resemble the target population. Synthetic covariates alone, however, are insufficient, because neither the RCT covariate distribution nor the DT covariate distribution necessarily matches the target covariate distribution. We therefore propose a statistical inference procedure that integrates RCTs with calibrated DTs using external target population summaries. The procedure represents the target covariate distribution as a calibrated mixture of the RCT and DT covariate distributions, allowing DTs to contribute covariate information while limiting the influence of poorly calibrated synthetic data. The procedure also incorporates LLM-generated auxiliary outcome predictions through calibrated outcome regressions, improving precision without changing the estimand. Theoretical results and simulation studies show that the proposed estimators reduce bias relative to RCT-only estimation and improve efficiency for overall and subgroup target treatment effects.

Large language model (LLM)-driven agents are proposed for a wide range of applications. As scenarios become increasingly composite, a graph-structured multi-agent system (MAS) offers a promising solution due to the orchestration of various skills. To architect a reasonable underlying workflow, recent advances mostly investigate two structures: sequential topology and decentralized topology. The former constrains the structure into a directed acyclic graph for multi-step executions but lacks sufficient scalability, while the latter leverages node-wise subgraphs for knowledge aggregation with synchronous parallelization but lacks long-term planning. Therefore, our key insight is that their strengths are mutually complementary. Driven by this motivation, we introduce step-aware hybrid topology for a MAS, where sequential connections support inter-step ordered planning and decentralized connections promote intra-step perspective integration. Concretely, to effectively coordinate different structural components, we clarify the corresponding topological definition and the protocols for step-wise serialization and parallelization. Subsequently, to dynamically refine desired structures, we introduce the parameterized distribution of a hybrid topology, which is initialized by LLM-driven graph planning. Eventually, to further reduce token consumption and time complexity, we supplement the topological density and length normalization for end-to-end learning. Extensive experiments demonstrate that the hybrid topology outperforms single-form topology in 90.4\% of tasks, our agentic initialization improves learning in 88.8\% of scenarios, proposed normalizations reduce token consumption by 20\%$\sim$30\% and time consumption by 7\%$\sim$13\% with a slight utility fluctuation. The code is available at https://anonymous.4open.science/r/HTAS-D107.

Large language models can hallucinate confidently, making uncertainty quantification essential for reliable deployment. Existing approaches rely predominantly on token-level signals from the final (unembedding) layer, while the rich geometric structure of intermediate hidden states remains largely unexplored. Aggregating token-level scores to the response level is itself non-trivial, and no single signal reliably catches the confident-but-wrong failure mode where a model commits to every token with high confidence yet produces an incorrect answer. We address this gap by extracting complementary signals from two distinct representational layers: hidden-state geometric complexity (global uncertainty) from the embedding layer, and token-level entropy (local uncertainty) from the unembedding layer. We show empirically that the two signals cover different regimes that are weakly correlated and thus combining them captures failure modes invisible to either alone. Building on this insight, we propose Global Local Uncertainty (GLU), an unsupervised, single-pass framework that fuses the two signals via a multiplicative gate. Experiments across three benchmarks and three model families show that GLU matches or outperforms all unsupervised baselines and remains competitive with supervised methods that lack cross-dataset generalization.

Expert background knowledge is often available in practical applications of causal discovery. Such constraints on the true causal graph can help causal discovery in terms of identifiability of causal effects and accuracy of the learned structure, but also in reducing the space of candidate causal graphs. As causal discovery can become computationally expensive for large number of variables, it is crucial to utilize background knowledge effectively \emph{during} the causal discovery process. However, most current methods only use background knowledge in a postprocessing step after causal discovery to refine the learned graph. In this work, we develop a framework for utilizing background knowledge during the causal discovery process, focusing especially on scalable causal discovery methods that recover only a subset of the whole graph. We implement our framework for multiple algorithms and empirically show that utilizing background knowledge can both reduce computational requirements and increase the quality of the learned structures.

Bounds based on Integral Probability Metrics (IPMs) are ubiquitous in machine learning and statistics, appearing, e.g., in generalization, alignment, fairness, and privacy. IPMs reduce the discrepancy between two distributions to a single worst-case expectation gap. This compression is costly: for rare events, safety failures, or under-represented groups, IPM-based bounds are loose or vacuous. We introduce a non-linear generalization of IPMs, which we term the Integral Probability Boundaries (IPBs). Unlike IPMs, IPBs provide bounds that are tight at the extremes. We derive basic properties of IPBs, and provide algorithms for estimating them in different practically relevant settings. We demonstrate the usage of IPBs in both theoretical and practical applications for alignment, judge-free performance estimation, and fairness, where they enable us to obtain sharp bounds on performance and fairness gaps, e.g., 15p.p. more precise estimates on fairness gaps for rare groups in a realistic generative modeling task, compared to state-of-the-art.