NeurIPS 2026 论文列表

High-performance inference for large models is critical in latency-sensitive applications. Existing inference engines suffer from sequential execution of hundreds of fine-grained kernels, causing kernel launch overheads and redundant memory accesses. A promising solution is fusing multiple kernels into a single large kernel, namely MegaKernel. However, existing MegaKernel implementations are tightly coupled to specific models and GPU architectures and require extensive manual engineering. We propose an LLM-driven framework AutoMegaKernel, which leverages an Instruction-Centric Fusion Abstraction to decompose MegaKernels into tractable fusion instructions, thereby enabling automated generation. Our framework jointly performs fusion strategy reasoning and CUDA code generation while hierarchically verifying correctness. Extensive experiments demonstrate that AutoMegaKernel consistently outperforms state-of-the-art inference engines (up to 5.04$\times$ vLLM and 2.42$\times$ SGLang) and compilers (up to 1.92$\times$ MPK), while supporting a broader range of models (dense, MoE, and VLA) and NVIDIA GPU platforms (Hopper, Ampere, and Ada).

Sign Language Video Generation (SLVG) aims to generate realistic and motion-accurate sign language videos from spoken text. Most existing studies heavily rely on intermediate representations (\eg poses or 3D meshes), and adopt a two-stage generation paradigm, where the text is first converted into poses or other visual modalities, which are then used as conditions to synthesize sign language video frames via diffusion models (\eg pose-to-image models). However, the end-to-end SLVG task remains largely unexplored. In this paper, we demonstrate that end-to-end SLVG is feasible without relying on intermediate representations(\eg poses), and we argue that a key challenge lies in how to generate effective and temporally consistent conditions. With well-designed condition modules, it becomes feasible to build an end-to-end SLVG model without relying on any intermediate modalities. To this end, we propose LLMCond, a first end-to-end text-to-video SLVG framework that requires no auxiliary visual cues during either training or inference. LLMCond consists of four key components: (1) a video VQVAE that compresses sign language videos into a sequence of discrete latent tokens; (2) \textbf{an LLM-based conditioner} that generates temporally and semantically aligned conditions corresponding to the sign sequences; (3) \textbf{a Gaussian Condition Refiner} designed to enforce temporal smoothness across conditions, enabling the generation of coherent and natural sign language motions; and (4) a discrete diffusion model that synthesizes motion-accurate sign language videos conditioned on the refined condition sequences. Extensive experiments on public SL datasets demonstrate that LLMCond achieves highly competitive performance, producing temporally coherent and accurate sign language motions without relying on auxiliary visual modalities such as pose, depth, or optical flow.

Video generation enables text-to-video synthesis, video editing, and motion-controlled content creation. However, current video generation models suffer from high computational latency, rendering true real-time capabilities infeasible for down stream tasks. We address this limitation by exploiting the temporal redundancy inherent in video latent patches. To this end, we propose the Latent Inter-frame Pruning with Attention Recovery (LIPAR) framework, which detects and skips recomputing duplicated latent patches. Additionally, we introduce a novel Attention Recovery mechanism that approximates the attention values of pruned tokens, thereby removing visual artifacts arising from naively applying the pruning method. Empirically, our method increases generation throughput by $1.45\times$, on average achieving 12.2 FPS on an NVIDIA A6000 compared to the baseline 8.4 FPS. The proposed method does not compromise generation quality and can be seamlessly integrated with Diffusion Transformer without additional training. Our approach effectively bridges the gap between traditional compression algorithms and modern generative pipelines.

Distilled diffusion models accelerate image generation by reducing the number of denoising steps, but often suffer from degraded image quality. To mitigate this trade-off, test-time optimization methods improve quality, yet their iterative nature incurs substantial computational overhead and leads to slow inference, limiting practical usability. Recent hypernetwork-based approaches amortize this process during training, but still require costly noise modulation in high-dimensional latent spaces. In this work, we propose LENS (Low-frequency Eigen Noise Shaping), an efficient noise modulation framework that operates in a low-dimensional subspace. Our approach is motivated by the observation that low-frequency components of the noise largely determine the global structure and visual fidelity of generated images. Based on this observation, we provide a theoretical justification for restricting modulation to the low-frequency subspace and derive a principled training objective. Building on this, LENS employs a lightweight, standalone network to selectively modulate these components, enabling efficient and targeted noise modulation. Extensive experiments demonstrate that LENS achieves competitive image quality while reducing FLOPs by 400–700$\times$, model parameters by 25–75$\times$, and inference-time overhead by 10–20$\times$ compared to prior methods.

Reinforcement learning (RL) has become central to post-training large language models, but its rollout generation stage is often the dominant system bottleneck. The inefficiency comes from two sources. First, autoregressive decoding makes latency grow with response length. Second, response lengths vary widely within a batch: short sequences finish early, but the batch remains blocked by the longest sequence, leaving GPU capacity underutilized. This effect is amplified in multi-sample rollouts, where long responses can also concentrate on a few decoding workers and delay synchronization. We propose LADDERS (Length-Aware Data Distribution and Existing-Response Speculation), a lightweight framework for accelerating RL rollout generation. LADDERS first uses hidden-state-based length prediction to group prompts with similar expected response lengths, and then applies an S-shaped allocation rule to balance multi-sample requests across workers. It further reuses responses already generated by the policy as draft continuations through prompt-specific suffix trees, enabling speculative decoding without an auxiliary draft model. Experiments on Qwen3-1.7B and Qwen3-8B show that LADDERS reduces rollout generation time by up to 56% while preserving final task performance, and can be integrated into existing RL systems with minimal engineering effort.

Training-free safeguards for text-to-image diffusion models often rely on a reusable safety signal, such as an unsafe direction or global toxic subspace, applied broadly across prompts. We provide a controlled geometric analysis of this global-unsafety assumption and reveal a consistent coverage-selectivity trade-off: compact unsafe subspaces fail to cover heterogeneous unsafe semantics, whereas broader aggregation increasingly distorts safety-adjacent benign prompts. Motivated by this finding, we propose CALM (Counterfactual Adaptive Local Modulation), a training-free safeguard that replaces uniform global removal with prompt-local counterfactual correction. Using matched unsafe-benign anchors, CALM routes each prompt to active unsafe categories, minimally edits only violating token representations toward the safe side, and suppresses positively aligned unsafe residual components. Across broad evaluation, CALM improves unsafe-content suppression while preserving benign utility, demonstrating that local counterfactual correction provides a more selective alternative to global unsafe-signal removal.

Neural posterior estimation has emerged as a powerful tool for amortized inference, with growing adoption across scientific and applied domains. In many of these applications, the conditioning variable is a set of observations whose elements depend not only on the target but also on unknown factors shared across the set. Optimal inference therefore requires treating the set jointly, which in turn requires training the estimator at the deployment set size---a regime where memory and compute quickly become prohibitive. We introduce a simple, theoretically grounded strategy that decouples representation learning from posterior modeling. Our method trains a mean-pool Deep Set on sets of size at most two, producing an encoder that generalizes to arbitrary set sizes. The inference head is then finetuned on pre-aggregated embeddings, making training cost essentially independent of the deployment set size $N$. Across scalar, image, multi-view 3D, molecular, and high-dimensional conditional generation benchmarks with $N$ in the thousands, our approach matches or outperforms standard baselines at a fraction of the compute.

Image-free zero-shot classifier expansion aims to extend recognition to unseen classes by establishing an association mapping between class semantics and visual classifier weights, without accessing original images. However, most existing methods rely on an association-based optimization mechanism, which struggles to meet the demands for fine-grained classification of visually and semantically similar classes. Since co-occurring attributes dominate class semantics, the model gravitates towards co-occurring attributes rather than discriminative attributes. This leads to geometrically entangled classifier weights, resulting in ambiguous decision boundaries. To address this problem, this paper proposes a novel approach that reframes weight generation through the lens of causal intervention. The method first employs comparative reasoning to disentangle discriminative and co-occurring attributes. Subsequently, it synthesizes intervened semantics to impose invariance and sensitivity constraints. This mechanism compels the model to suppress co-occurring attributes while enhancing sensitivity to discriminative attributes. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches, yielding clearer decision boundaries in fine-grained settings.

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.

Text-to-image (T2I) models are increasingly embedded in everyday visual culture, yet little is known about how they translate abstract human values into visual scenes. We address this gap by introducing a framework, grounded in Schwartz’s theory of basic human values, for studying value instantiation: the process by which ideals such as success, care, or freedom are visually realized. Using person-centered prompts from 70 positive and negative value items, we generate 28,000 images across four state-of-the-art T2I models. We identify visual prototypes, recurring combinations of subjects, actions, and settings, and use them to quantify how narrowly models instantiate these values and how this breadth compares with human descriptions of the same items. We then map these prototypes onto a taxonomy of social configurations spanning setting, sociality, mode of engagement, and social register, revealing what is systematically foregrounded or omitted. Our results show that T2I models produce far more concentrated representations than humans do when describing the same values. They default to a hyper-individualistic, adult-centric worldview, reproducing value-specific associations, such as achievement tied to formal institutional settings or hedonism rendered as affluent consumption, while marginalizing everyday labor, relational care, and non-adult perspectives. As a proof of concept, we show that these structural gaps can guide a steering method that diversifies outputs without modifying the original prompt.

Instruction-based image editing exhibits heterogeneous difficulty not only across cases but also across regions of an image, motivating refinement approaches that allocate correction to where the model struggles. Existing refinement signals arrive late, after a fully generated image or a completed denoising step. We ask whether such a signal can act *within* an ongoing forward pass. To investigate this, we probe a frozen image-editing model and find that although generation capability emerges only in the last few layers, the error pattern is already set in early layerss (rank correlation ρ = 0.83 with the final-layer error map). Based on this, we introduce **Inline Critic**, a learnable token that critiques a frozen model's predictions at its intermediate layers and steers its hidden states to refine generation during the forward pass. A three-stage recipe is proposed to stabilize the training from learning how to critique to steering generation. As a result, we achieve state of the art on GEdit-Bench (7.89), a +9.4 gain on RISEBench over the same backbone, and the strongest open-source result on KRIS-Bench (81.92, surpassing GPT-4o). We further provide analyses showing that the critic genuinely shapes the model's attention and prediction updates at subsequent layers.

Safety-aligned Large Language Models (LLMs) remain vulnerable to interventions during inference that redirect generation toward harmful outputs. Recent work attributes this to shallow safety, where alignment concentrates in the first few output tokens. We show that shallow safety is a special case of a broader inference-time vulnerability, in which short token injections at any generation step can substantially alter subsequent safety behavior. We also find that a model's alignment with refusal directions in its hidden states does not predict its robustness to such injection, revealing that internal state alone does not determine generation behavior under perturbation. To address this, we align models directly on generation trajectories constructed by simulating mid-sequence perturbation, and show that this improves robustness to mid-sequence injection and generalizes to attacks that exploit early-token generation. Our work argues that robust safety alignment requires training on the generation process itself, not only its outputs.

Code generation and comprehension by Large Language Models (LLMs) have emerged as core drivers of industrial intelligence and decision optimization, finding widespread application in fields such as finance, automation, and aerospace. Although recent advancements have demonstrated the remarkable potential of LLMs in general code generation, existing benchmarks are mainly confined to single domains and languages. Consequently, they fail to effectively evaluate the generalization capabilities required for real-world industrial applications or to reflect the coding proficiency demanded by complex industrial scenarios. To bridge this gap, we introduce IndustryCode, the first comprehensive benchmark designed to span multiple industrial domains and programming languages. IndustryCode comprises 579 sub-problems derived from 125 primary industrial challenges, accompanied by rigorous problem descriptions and test cases. It covers a wide range of NAICS sectors, including Aerospace Product Mfg, Chemical Mfg, and Information, and incorporates diverse programming languages such as MATLAB, Python, C++, and Stata. In our evaluation, the top-performing model, Claude 4.7 Opus, achieved an overall accuracy of 71.9\% on sub-problems and 48.9\% on main problems. The benchmark dataset and automated evaluation code are now publicly available.

Classifier-free guidance (CFG) is essential for improving sample quality in visual generation, but it incurs substantial sampling overhead. Existing guidance-free approaches halve this cost by approximating the CFG with a single model inference, but their performance is bounded by the CFG and inherits its limitations. In this paper, we propose self-contrastive guidance (SCG), which contrasts the model's prediction on real and self-generated samples to obtain a direct, model-based signal that pushes generation toward the true data manifold. Integrating SCG into the standard likelihood objective via reparameterization yields Self-Contrastive Alignment for Likelihood Estimation (SCALE), a guidance-free training framework that produces SCG-enhanced predictions in a single forward pass. Across class-conditional and text-to-image generation on diffusion and autoregressive models, SCALE consistently outperforms CFG-mimicking guidance-free baselines and improves further when combined with DPO, demonstrating that the self-contrastive signal yields gains complementary to both CFG and preference-based fine-tuning.

Diffusion-based posterior sampling (PS) is a leading framework for imaging inverse problems, combining learned priors with measurement constraints. Yet, its standard formulations rely on instantaneous data-consistent estimates, which induce temporal variability in the reverse dynamics. We reinterpret PS from a dynamical perspective, showing that the standard PS update corresponds to a first-order discretization of the diffusion dynamics plus a residual correction capturing the mismatch between the denoised prediction and the data-consistent estimate. A second-order discretization, however, naturally introduces a temporal correction based on the variation of consecutive estimates. Building on this, we propose LAMP, combining the second-order update with the residual correction characterizing a PS technique. LAMP thus inherits a lagged temporal correction, and it can be implemented as a modular plug-in over the PS backbone. We show that LAMP preserves the structure of a posterior sampler, and we perform a one-step risk analysis to characterize when LAMP improves the reverse transition via a bias-variance trade-off. Experiments across multiple imaging tasks demonstrate consistent improvements over strong baselines such as DiffPIR and DDRM, without increasing the number of denoising evaluations.

Incorporating visual semantic representations as an intermediate step before image generation can reduce the modeling difficulty between text and images, thereby improving generation quality. Recent works such as X-Omni and BLIP3o-Next have explored this direction, but they typically use a two-stage external pipeline: a separate autoregressive model first generates semantic tokens, which are then fed as conditioning to an independent diffusion decoder. Since the decoder cannot jointly access the original input and the semantic plan, this design introduces an information bottleneck that limits detail preservation in downstream tasks such as editing. Internal architectures such as Transfusion, BAGEL, and Show-o2 avoid this bottleneck by enabling cross-modal interaction within a single model, but they still face the difficult text-to-pixel modeling gap without intermediate semantic guidance. We propose Visual Prompt Engineering (VPE), which can be seamlessly integrated into such internal frameworks. Specifically, the model first autoregressively generates visual semantic tokens ($e.g.$, SigLIP 2) as "visual prompts" that capture the semantic layout, then generates the full image tokens conditioned on this plan. We validate VPE across class-conditional generation, text-to-image generation, and image editing, covering various token types and model architectures. Results show that VPE can accelerate convergence, raise quality ceilings, and through internal integration, achieve substantially better editing preservation (PSNR: $26.76$ vs. $19.92$) than external alternatives of the same parameter scale, while maintaining competitive editing responsiveness. The code is available in supplementary material.

ImageNet class-conditional generation is a standard first test for new image generation methods. Progress on ImageNet is usually measured with one number: Fréchet Inception Distance (FID), computed by comparing 50K class-balanced generated samples with the ImageNet training set in the Inception-v3 feature space. The FID number was useful when the generation quality was low and the gap between methods was large. However, many recent methods already reach remarkable and close FID numbers, and we argue that the field should stop treating FID as decisive evidence of generation quality. To support this view, we re-evaluate 47 recent ImageNet class-conditional generators while keeping the generated samples fixed and changing only the evaluation protocol. We test multiple variants of FID, including different real reference sets, sample sizes, and feature extractors such as Inception-v3, CLIP, DINOv2, and VGG16. The same generated samples can receive different scores, resulting in varying rankings. Our observation means that a single FID protocol has a risk of being over-optimized and over-interpreted. We do not argue that FID should be removed. Instead, we recomend: keep the conventional ImageNet FID for continuity, but also report the results of some FID varaints to avoid overfitting to one evaluation recipe. If a method reaches comparable and stable ImageNet FID, authors should not be expected to keep chasing the best FID, and the paper should be evaluated based on their technical contributions. After passing the FID test, future work such as text-to-image training and human preference study, is important for the futher development of a generation method.

Modern LLM APIs often reveal only top-$K$ logit scores and censor the remaining vocabulary. We study the per-position distribution-recovery limits of this access model. For censoring threshold $\tau$, the compatible teacher distributions form an identified set whose total-variation diameter is exactly $U_K=(V{-}K)\exp(\tau)/(Z_A+(V{-}K)\exp(\tau)),$ where $Z_A$ is the observed partition function. For KL recovery, we give a computable binary-endpoint lower bound and an asymptotically matching small-ambiguity upper bound, with an extension to reference-aware attackers. Experiments on a Qwen3 math-reasoning teacher reveal a layered extraction hierarchy: on-task top-$K$ distillation recovers 12% of private capability, full-logit distillation recovers 56% despite 99% KL closure, and generation-based extraction recovers 96%. Top-$K$ censoring therefore limits per-position distribution recovery but does not by itself prevent capability extraction, separating fidelity from transfer in prompt-only logit distillation.

Feature caching accelerates Diffusion Transformers by reusing intermediate computations across denoising timesteps, yet existing methods share a fundamental blind spot: they treat the standard reference denoising trajectory as the optimisation target, approximating its per-step states and thus inheriting its output as an implicit quality ceiling. We argue that this target is arbitrary --- what truly matters is the ideal result, not the intermediate path taken to reach it. We propose IdealCache, a training-free framework that abandons step-wise imitation of the reference trajectory. Instead, it anchors caching on a per-input ensemble of ideal trajectories constructed from candidate target results, aggregates their geometry into a single graph, and discovers optimal non-uniform schedules via a constrained shortest path on this graph. By decoupling from the reference process, IdealCache breaks the performance bottleneck of trajectory-approximation; our schedules can attain near-lossless fidelity in generation and even surpass on task-specific metrics (e.g., maintaining higher semantic consistency in image editing). Without retraining, IdealCache achieves $6.35\times$ speedup on Qwen-Image, $5.84\times$ on Qwen-Image-Edit, and $4.32\times$ on HunyuanVideo, with competitive or superior quality across tasks.