论文检索

Deep Learning · Generative Models and Autoencoders

Candi Zheng, Yuan Lan

Diffusion models are often introduced from multiple perspectives—such as VAEs, score matching, or flow matching—accompanied by dense and technically demanding mathematics that can be difficult for beginners to grasp. This article offers a fresh Langevin perspective on diffusion models to lower the technical barrier, aiming to present diffusion models in a simpler, clearer, and more intuitive way while addressing the following questions: 1. How does the reverse process invert the forward process to generate data from pure noise? 2. How can ODE-based and SDE-based diffusion models be unified under a single framework? 3. Why are diffusion models theoretically superior to ordinary VAEs? 4. How can Denoising, Score Matching, and Flow Matching training objectives be unified and derived from first principles? We demonstrate that the Langevin perspective offers clear and straightforward answers to these questions, providing pedagogical value for both learners and experienced researchers seeking deeper intuition.

Applications · Chemistry and Drug Discovery

Richard Strunk, Karnik Ram, Daniel Cremers

MLIPs are a promising paradigm in atomistic simulation, potentially offering the accuracy of ab-initio methods at the speed of empirical potentials. In this blog post, we give an overview of recent MLIP architectures, followed by an evaluation on a practical CO2 adsorption simulation. We find that as of today these models, though promising, are far from plug-and-play, requiring significant engineering effort to operate within established simulation frameworks, while also failing to produce physically consistent results.

Deep Learning · Generative Models and Autoencoders

Mingyu Kim, Young-Heon Kim, Mijung Park

Safety mechanisms for diffusion and flow models have recently been developed along two distinct paths. In robot planning, control barrier functions are employed to guide generative trajectories away from obstacles at every denoising step by explicitly imposing geometric constraints. In parallel, recent data-driven, negative guidance approaches have been shown to suppress harmful content and promote diversity in generated samples. However, they rely on heuristics without clearly stating when safety guidance is actually necessary. In this paper, we first introduce a unified probabilistic framework using a Maximum Mean Discrepancy (MMD) potential for image generation tasks that recasts both Shielded Diffusion and Safe Denoiser as instances of our energy-based negative guidance against unsafe data samples. Furthermore, we leverage control-barrier functions analysis to justify the existence of a critical time window in which negative guidance must be strong; outside of this window, the guidance should decay to zero to ensure safe and high-quality generation. We evaluate our unified framework on several realistic safe generation scenarios, confirming that negative guidance should be applied in the early stages of the denoising process for successful safe generation.

Computer Vision · Vision Models & Multimodal

Minyoung Lee, Yeji Park, Dongjun Hwang, Yejin Kim, Seong Joon Oh, Junsuk Choe

Large Vision-Language Models (LVLMs) achieve strong performance on single-image tasks, but their performance declines when multiple images are provided as input. One major reason is the cross-image information leakage, where the model struggles to distinguish information across different images. Existing LVLMs already employ delimiter tokens to mark the start and end of each image, yet our analysis reveals that these tokens fail to effectively block cross-image information leakage. To enhance their effectiveness, we propose a method that scales the hidden states of delimiter tokens. This enhances the model’s ability to preserve image-specific information by reinforcing intra-image interaction and limiting undesired cross-image interactions. Consequently, the model is better able to distinguish between images and reason over them more accurately. Experiments show performance gains on multi-image benchmarks such as Mantis, MuirBench, MIRB and QBench2. We further evaluate our method on text-only tasks that require clear distinction. The method improves performance on multi-document and multi-table understanding benchmarks, including TQABench, MultiNews and WCEP-10. Notably, our method requires no additional training or inference cost.

Applications · Everything Else

Moises Andrade, Joonhyuk Cha, Brandon Ho, Vriksha Srihari, Karmesh Yadav, Zsolt Kira

Verifiers—functions assigning rewards to agent behavior—have been key to AI progress in domains such as math, code, and games. However, extending these gains to domains without clear-cut success criteria (e.g., computer use) remains a challenge: while humans can recognize desired outcomes, translating this intuition into scalable rules is nontrivial. Multimodal LLMs (MLLMs) emerge as a promising solution, given vast world knowledge, human-preference alignment, and reasoning capabilities. We evaluate MLLMs as verifiers across web navigation, computer use, and robotics, spanning 13+ model families, 28+ evaluation templates, curated trajectories from diverse agents and of varying lengths, and distinct verifier applications. We identify a critical limitation: a strong tendency for MLLMs to over-validate agent behavior—a phenomenon we term agreement bias. This bias is pervasive across models, resilient to test-time scaling, and can harm methods relying on MLLM evaluations, such as filtered behavior cloning and self-improvement. We provide guidance on the design and evaluation of MLLM verifiers, and introduce Self-Grounded Verification (SGV), a lightweight method that harnesses MLLMs' own sampling mechanisms by modulating (un)conditional generation to better leverage their knowledge, alignment, and reasoning. SGV operates in two steps: first, the MLLM is elicited to generate broad priors about desired behavior, independent of the data under evaluation. Then, conditioned on self-generated priors, it reasons over and evaluates a candidate trajectory. Our methods yield gains across models and environments, improving failure detection by up to 25pp and accuracy by 14pp, with benefits extending to downstream applications. In self-improvement and online supervision, SGV boosts task completion of a GUI specialist in OSWorld, a diffusion policy in robomimic, and a ReAct agent in VisualWebArena—setting a new state of the art, surpassing the previous best by 20pp. Finally, we release an updated version of VisualWebArena featuring strong agent baselines, more human-aligned evaluators, high-fidelity environment parallelism, runtime speedups exceeding 10x, and VisualWebArena-Lite, a 1/3-scale subset with comparable evaluation fidelity. Our code, models, and data are publicly available at [our project page](https://mshalimay.github.io/agreement-bias-sgv/).

Social Aspects · Fairness, Equity, Justice and Safety

Wen Huang, Jiarui Yang, Tao Dai, Jiawei Li, Shaoxiong Zhan, Bin Wang, Shu-Tao Xia

Visual manipulation localization (VML) aims to identify tampered regions in images and videos, a task that has become increasingly challenging with the rise of advanced editing tools. Existing methods face two central issues. The first is resolution diversity. Resizing or padding can distort subtle forensic cues and introduce unnecessary computational cost. The second is the difficulty of extending spatial models for images to spatio-temporal inputs in videos, which often results in maintaining separate architectures for the two data types. To address these challenges, we propose RelayFormer, a unified framework that adapts to varying resolutions and naturally handles both static and temporal visual data. RelayFormer partitions inputs into fixed-size sub-images and introduces Global Local Relay (GLR) tokens that propagate structured context through a relay-based attention mechanism. This design enables efficient exchange of global cues, such as semantic or temporal consistency, while preserving fine-grained manipulation artifacts. Unlike prior approaches that depend on uniform resizing or sparse attention, RelayFormer scales to variable resolutions and video sequences with minimal overhead. Experiments across diverse benchmarks demonstrate superior performance and strong efficiency, combining resolution adaptivity without interpolation or excessive padding, unified processing for images and videos, and a favorable balance between accuracy and computational cost. Code is available at~\href{https://github.com/WenOOI/RelayFormer}{https://github.com/WenOOI/RelayFormer}.

Computer Vision · Vision Models & Multimodal

Hyeonjun Jeong, Juyeb Shin, Dongsuk Kum

Neural radiance fields (NeRFs) have emerged as a prominent pre-training paradigm for vision-centric autonomous driving, which enhances 3D geometry and appearance understanding in a fully self-supervised manner. To apply NeRF-based pre-training to 3D perception models, recent approaches have simply applied NeRFs to volumetric features obtained from view transformation. However, coupling NeRFs with view transformation inherits conflicting priors; view transformation imposes discrete and rigid representations, whereas radiance fields assume continuous and adaptive functions. When these opposing assumptions are forced into a single pipeline, the misalignment surfaces as blurry and ambiguous 3D representations that ultimately limit 3D scene understanding. Moreover, the NeRF network for pre-training is discarded during downstream tasks, resulting in inefficient utilization of enhanced 3D representations through NeRF. In this paper, we propose a novel NeRF-Resembled Point-based 3D detector that can learn continuous 3D representation and thus avoid the misaligned priors from view transformation. NeRP3D preserves the pre-trained NeRF network regardless of the tasks, inheriting the principle of continuous 3D representation learning and leading to greater potentials for both scene reconstruction and detection tasks. Experiments on nuScenes dataset demonstrate that our proposed approach significantly improves previous state-of-the-art methods, outperforming not only pretext scene reconstruction tasks but also downstream detection tasks.

Deep Learning · Generative Models and Autoencoders

Kartik Sharma, Rakshit Trivedi

Activation steering methods enable inference-time control of large language model (LLM) behavior without retraining, but current approaches face a fundamental trade-off: sample-efficient methods suboptimally capture steering signals from labeled examples, while methods that better extract these signals require hundreds to thousands of examples. We introduce COLD-Steer, a training-free framework that steers LLM activations by approximating the representational changes that would result from gradient descent on in-context examples. Our key insight is that the effect of fine-tuning on a small set of examples can be efficiently approximated at inference time without actual parameter updates. We formalize this through two complementary approaches: (i) a unit kernel approximation method that updates the activations directly using gradients with respect to them, normalized across examples, and (ii) a finite-difference approximation requiring only two forward passes regardless of example count. Experiments across a variety of steering tasks and benchmarks demonstrate that COLD-Steer achieves upto 95\% steering effectiveness while using 50 times fewer samples compared to the best baseline. COLD-Steer enables real-time adaptation to new steering objectives and facilitates accommodating diverse perspectives without extensive demonstration data, which we validate through our experiments on pluralistic alignment tasks. Our framework opens new possibilities for adaptive, context-aware model control that can flexibly address varying loss-driven human preferences through principled approximation of learning dynamics rather than specialized training procedures.

General Machine Learning · Probabilistic Methods

Eshant English, Christoph Lippert

Conformal prediction provides a model-agnostic framework for uncertainty quantification with finite-sample validity guarantees, making it an attractive tool for constructing reliable prediction sets. However, existing approaches commonly rely on residual-based conformity scores, which impose geometric constraints and struggle when the underlying distribution is multimodal. In particular, they tend to produce overly conservative prediction areas centred around the mean, often failing to capture the true shape of complex predictive distributions. In this work, we introduce JAPAN (Joint Adaptive Prediction Areas with Normalising-Flows), a flow-based framework that uses density estimates for several conformal scores. By leveraging flow-based models, JAPAN estimates the (predictive) density and constructs prediction areas by thresholding on the estimated density scores, enabling compact, potentially disjoint, and context-adaptive regions that retain finite-sample coverage guarantees. We theoretically motivate the efficiency of JAPAN and empirically validate it across multivariate regression and forecasting tasks, demonstrating good calibration and tighter prediction areas compared to existing baselines. Furthermore, several density-based conformity scores showcase the flexibility of our proposed framework.

Reinforcement Learning · Deep RL

Runzhe Zhan, Yafu Li, Zhi Wang, Xiaoye Qu, Dongrui Liu, Jing Shao, Derek Wong, Yu Cheng

Reinforcement learning from verifiable rewards (RLVR) is an emerging paradigm for improving the reasoning ability of large language models. However, standard on-policy training discards rollout experiences after a single update, leading to computational inefficiency and instability. While prior work on RL has highlighted the benefits of reusing past experience, the role of experience characteristics in shaping learning dynamics of large reasoning models remains underexplored. In this paper, we are the first to investigate what makes a reasoning experience valuable and identify rollout correctness and entropy as effective indicators of experience value. Based on these insights, we propose ExGRPO (Experiential Group Relative Policy Optimization), a framework that organizes and prioritizes valuable experiences, and employs a mixed-policy objective to balance exploration with experience exploitation. Experiments on five backbone models (1.5B-8B parameters) show that ExGRPO consistently improves reasoning performance on mathematical/general benchmarks, with an average gain of +3.5/7.6 points over on-policy RLVR. Moreover, ExGRPO stabilizes training on both stronger and weaker models where on-policy methods fail. These results highlight principled experience management as a key ingredient for efficient and scalable RLVR.

General Machine Learning · Everything Else

Zihuan Qiu, Lei Wang, Yang Cao, Runtong ZHANG, Bing Su, Yi Xu, Fanman Meng, Linfeng Xu, Qingbo Wu, Hongliang Li

Data-free continual model merging (DFCMM) aims to fuse independently fine-tuned models into a single backbone that evolves with incoming tasks without accessing task data. This paper revisits two fundamental desiderata for DFCMM: stability, avoiding interference with earlier tasks, and plasticity, adapting faithfully to each new task. This poses a challenge that existing approaches fail to address: how to bridge data-level desiderata with parameter-space optimization to ensure stability and plasticity in the absence of task data. To this end, we propose NUFILT(NUll-space FILTering) a data-free framework that directly links these desiderata into parameter-space optimization. Our key observation is that task vectors approximately align with representation subspaces, providing structural surrogates for enforcing stability and plasticity. Accordingly, we design a null-space projector that preserves prior responses by filtering overlapping components of new task vectors, ensuring stability. We further introduce a lightweight LoRA adapter that injects complementary task-specific signals to enable plasticity. The adapter is trained with a projection-based surrogate loss that preserves consistency with prior knowledge while introducing novel directions. This joint filtering–adaptation process enables the backbone to absorb new knowledge while retaining existing behaviors, with updates fused back in a layer-wise linear fashion without extra parameters or inference cost. Theoretically, we establish approximate subspace alignment guarantees that justify null-space filtering. Empirically, NUFILT achieves state-of-the-art performance with minimal forgetting on both vision and NLP benchmarks, improving average accuracy by 4–7% over OPCM and WUDI-Merging, while narrowing the gap to fine-tuning and reducing computation overhead. The code is available at: https://github.com/zihuanqiu/NUFILT

General Machine Learning · Representation Learning

Yufei Zhu, Shih-Min Yang, Andrey Rudenko, Tomasz Kucner, Achim Lilienthal, Martin Magnusson

Safe and efficient robot operation in complex human environments can benefit from good models of site-specific motion patterns. Maps of Dynamics (MoDs) provide such models by encoding statistical motion patterns in a map, but existing representations use discrete spatial sampling and typically require costly offline construction. We propose a continuous spatio-temporal MoD representation based on implicit neural functions that directly map coordinates to the parameters of a Semi-Wrapped Gaussian Mixture Model. This removes the need for discretization and imputation for unevenly sampled regions, enabling smooth generalization across both space and time. Evaluated on two public datasets with real-world people tracking data, our method achieves better accuracy of motion representation and smoother velocity distributions in sparse regions while still being computationally efficient, compared to available baselines. The proposed approach demonstrates a powerful and efficient way of modeling complex human motion patterns and high performance in the trajectory prediction downstream task. The code is publicly available at https://github.com/test-bai-cpu/nemo-map.

Jianxin Zhang, Clayton Scott

Many generative models originally developed in finite-dimensional Euclidean space have functional generalizations in infinite-dimensional settings. However, the extension of rectified flow to infinite-dimensional spaces remains unexplored. In this work, we establish a rigorous functional formulation of rectified flow in an infinite-dimensional Hilbert space. Our approach builds upon the superposition principle for continuity equations in an infinite-dimensional space. We further show that this framework extends naturally to functional flow matching and functional probability flow ODEs, interpreting them as nonlinear generalizations of rectified flow. Notably, our extension to functional flow matching removes the restrictive measure-theoretic assumptions in the existing theory of \citet{kerrigan2024functional}. Furthermore, we demonstrate experimentally that our method achieves superior performance compared to existing functional generative models.

Theory · Learning Theory

Peter Shaw, James Cohan, Jacob Eisenstein, Kristina Toutanova

The Minimum Description Length (MDL) principle offers a formal framework for applying Occam's razor in machine learning. However, its application to neural networks such as Transformers is challenging due to the lack of a principled, universal measure for model complexity. This paper introduces the theoretical notion of asymptotically optimal description length objectives, grounded in the theory of Kolmogorov complexity. We establish that a minimizer of such an objective achieves optimal compression, for any dataset, up to an additive constant, in the limit as model resource bounds increase. We prove that asymptotically optimal objectives exist for Transformers, building on a new demonstration of their computational universality. We further show that such objectives can be tractable and differentiable by constructing and analyzing a variational objective based on an adaptive Gaussian mixture prior. Our empirical analysis shows that this variational objective selects for a low-complexity solution with strong generalization on an algorithmic task, but standard optimizers fail to find such solutions from a random initialization, highlighting key optimization challenges. More broadly, by providing a theoretical framework for identifying description length objectives with strong asymptotic guarantees, we outline a potential path towards training neural networks that achieve greater compression and generalization.

Deep Learning · Everything Else

Shihao Zhang, Haoyu Zhang, Ian Colbert, Rayan Saab

We introduce Qronos---a new post-training quantization algorithm that not only explicitly corrects errors due to both weight and activation quantization, but also corrects errors accumulated from previously quantized layers. Our iterative algorithm is based on an interpretable and disciplined optimization framework that surpasses existing data-driven approaches. At each step, Qronos alternates between error correction and diffusion via optimal update rules. Importantly, we prove that Qronos admits an equivalent formulation that significantly improves algorithmic efficiency; we use our discovery to reduce peak memory usage by 18\times on Llama3 8B, and our scaling analysis shows a speedup of up to 13.8\times for a single-layer microbenchmark. We demonstrate compatibility with existing transformation techniques such as Hadamard-based incoherence processing and weight-activation scaling equalization, among others. We evaluate Qronos using recent language models in the Llama3 and Qwen3 families; Qronos consistently outperforms previous state-of-the-art adaptive rounding methods when quantizing the weights, activations, and/or KV caches to 4 bits or fewer.

Applications · Language, Speech and Dialog

Gregory Yauney, Shahzaib Warraich, Swabha Swayamdipta

Micro-benchmarking offers a solution to the often prohibitive time and cost of language model development: evaluate on a very small subset of existing benchmarks. Can these micro-benchmarks, however, rank models as consistently as the full benchmarks they replace? And can they rank models more consistently than selecting a random subset of data points? In many scenarios, we find that the answer is no. We introduce a meta-evaluation measure for micro-benchmarking which investigates how well a micro-benchmark can rank two models as a function of their performance difference on the full benchmark. This approach can determine which model pairs can be ranked correctly by a micro-benchmark, allowing for a finer-grained analysis of the trade-off between micro-benchmark size and reliability. Prior work has suggested selecting as few as 10 examples; we find that no micro-benchmarking method can consistently rank model pairs 3.5 points of accuracy apart on MMLU-Pro or 4 points apart on BIG-bench Hard. In order to consistently rank model pairs with relatively similar performances, we show that often as many as 250 examples must be selected, at which point random sampling is competitive with existing micro-benchmarking methods. When comparing only 8B instruction-tuned models on MMLU-Pro micro-benchmarks with 25 examples, we find that more than half of pairwise comparisons are not likely to be preserved. Our work provides actionable guidance for both micro-benchmark users and developers in navigating the trade-off between evaluation efficiency and reliability.

Reinforcement Learning · Everything Else

Bill Zheng, Vivek Myers, Benjamin Eysenbach, Sergey Levine

The problem of learning how to reach goals in an environment has been a long- standing challenge in for AI researchers. Effective goal-conditioned reinforcement learning (GCRL) methods promise to enable reaching distant goals without task- specific rewards by stitching together past experiences of different complexity. Mathematically, there is a duality between the notion of optimal goal-reaching value functions (the likelihood of success at reaching a goal) and temporal dis- tances (transit times states). Recent works have exploited this property by learning quasimetric distance representations that stitch long-horizon behaviors using the in- ductive bias of their architecture. These methods have shown promise in simulated benchmarks, reducing value learning to a shortest-path problem. But quasimet- ric, and more generally, goal-conditioned RL methods still struggle in complex environments with stochasticity and high-dimensional (visual) observations. There is a fundamental tension between the local dynamic programming (TD backups, temporal distances) that enables optimal shortest-path reasoning in theory and the statistical global MC updates (multistep returns, suboptimal in theory). We show how these approaches can be integrated into a practical GCRL method that fits a quasimetric distance using a multistep Monte-Carlo return. We show our method outperforms existing GCRL methods on long-horizon simulated tasks with up to 4000 steps, even with visual observations. We also demonstrate that our method can enable stitching in the real-world robotic manipulation domain (Bridge setup). Our approach is the first end-to-end GCRL method that enables multistep stitching in this real-world manipulation domain from an unlabeled offline dataset of visual observations.

Deep Learning · Theory

Adel Javanmard, Baharan Mirzasoleiman, Vahab Mirrokni

Test-time scaling improves the reasoning capabilities of large language models (LLMs) by allocating extra compute to generate longer Chains-of-Thoughts (CoTs). This enables models to tackle more complex problem by breaking them down into additional steps, backtracking, and correcting mistakes. Despite its strong performance--demonstrated by OpenAI's o1 and DeepSeek R1, the conditions in the training data under which long CoTs emerge, and when such long CoTs improve the performance, remain unclear. In this paper, we study the performance of test-time scaling for transformers trained on an in-context weight prediction task for linear regression. Our analysis provides a theoretical explanation for several intriguing observations: First, at any fixed test error, increasing test-time compute allows us to reduce the number of in-context examples (context length) in training prompts. Second, if the skills required to solve a downstream task are not sufficiently present in the training data, increasing test-time compute can harm performance. Finally, we characterize task hardness via the smallest eigenvalue of its feature covariance matrix and show that training on a diverse, relevant, and hard set of tasks results in best performance for test-time scaling. We confirm our findings with experiments on large, nonlinear transformer architectures.

General Machine Learning · Transfer, Multitask and Meta-learning

Gregory De Ruyter, Mathias Verbeke, Hans Hallez

Neural Architecture Search (NAS), particularly gradient-based techniques, has proven highly effective in automating the design of neural networks. Recent work has extended NAS to hardware-aware settings, aiming to discover architectures that are both accurate and computationally efficient. Many existing methods integrate hardware metrics into the optimization objective as regularization terms, which introduces differentiability requirements and hyperparameter tuning challenges. This can either result in overly penalizing resource-intensive architectures or architectures failing to meet the hardware constraints of the target device. To address these challenges, we propose ConNAS, a novel gradient-based NAS framework that enforces hardware constraints directly through gradient modification. This approach eliminates the need for differentiable hardware metrics and regularization weights. The novelty in ConNAS lies in modifying gradients with respect to architectural choices, steering the search away from infeasible architectures while ensuring constraint satisfaction. Evaluations on the NATS-Bench benchmark demonstrate that ConNAS consistently discovers architectures that meet the imposed hardware constraints while achieving performance within just 0.14% of the optimal feasible architecture. Additionally, in a practical deployment scenario, ConNAS outperforms handcrafted architectures by up to 1.55% in accuracy under tight hardware budgets. Our code is publicly available at https://gitlab.kuleuven.be/m-group-campus-brugge/distrinet_public/connas.

Computer Vision · Image and Video Generation

Yixuan Zhu, Shilin Ma, Haolin Wang, Ao Li, Yanzhe Jing, Yansong Tang, Lei Chen, Jiwen Lu, Jie Zhou

Recent advancements in visual autoregressive models (VAR) have demonstrated their effectiveness in image generation, highlighting their potential for real-world image super-resolution (Real-ISR). However, adapting VAR for ISR presents critical challenges. The next-scale prediction mechanism, constrained by casual attention, fails to fully exploit global low-quality (LQ) context, resulting in blurry and inconsistent high-quality (HQ) outputs. Additionally, error accumulation in the iterative prediction severely degrades coherence in ISR task. To address these issues, we propose VARestorer, a simple yet effective distillation framework that transforms a pre-trained text-to-image VAR model into a one-step ISR model. By leveraging distribution matching, our method eliminates the need for iterative refinement, significantly reducing error propagation and inference time. Furthermore, we introduce pyramid image conditioning with cross-scale attention, which enables bidirectional scale-wise interactions and fully utilizes the input image information while adapting to the autoregressive mechanism. This prevents later LQ tokens from being overlooked in the transformer. By fine-tuning only 1.2\% of the model parameters through parameter-efficient adapters, our method maintains the expressive power of the original VAR model while significantly enhancing efficiency. Extensive experiments show that VARestorer achieves state-of-the-art performance with 72.32 MUSIQ and 0.7669 CLIPIQA on DIV2K dataset, while accelerating inference by 10 times compared to conventional VAR inference.