ICLR 2025 论文列表

The blooming of virtual reality and augmented reality (VR/AR) technologies has driven an increasing demand for the creation of high-quality, immersive, and dynamic environments. However, existing generative techniques either focus solely on dynamic objects or perform outpainting from a single perspective image, failing to meet the requirements of VR/AR applications that need free-viewpoint, 360$^{\circ}$ virtual views where users can move in all directions. In this work, we tackle the challenging task of elevating a single panorama to an immersive 4D experience. For the first time, we demonstrate the capability to generate omnidirectional dynamic scenes with 360$^{\circ}$ views at 4K (4096 $\times$ 2048) resolution, thereby providing an immersive user experience. Our method introduces a pipeline that facilitates natural scene animations and optimizes a set of 3D Gaussians using efficient splatting techniques for real-time exploration. To overcome the lack of scene-scale annotated 4D data and models, especially in panoramic formats, we propose a novel Panoramic Denoiser that adapts generic 2D diffusion priors to animate consistently in 360$^{\circ}$ images, transforming them into panoramic videos with dynamic scenes at targeted regions. Subsequently, we propose Dynamic Panoramic Lifting to elevate the panoramic video into a 4D immersive environment while preserving spatial and temporal consistency. By transferring prior knowledge from 2D models in the perspective domain to the panoramic domain and the 4D lifting with spatial appearance and geometry regularization, we achieve high-quality Panorama-to-4D generation at a resolution of 4K for the first time. Project page: https://4k4dgen.github.io/.

The development of large language models (LLMs) has significantly advanced the emergence of large multimodal models (LMMs). While LMMs have achieved tremendous success by promoting the synergy between multimodal comprehension and creation, they often face challenges when confronted with out-of-distribution data, such as which can hardly distinguish orientation, quantity, color, structure, etc. This is primarily due to their reliance on image encoders trained to encode images into task-relevant features, which may lead them to disregard irrelevant details. Delving into the modeling capabilities of diffusion models for images naturally prompts the question: Can diffusion models serve as the eyes of large language models for image perception? In this paper, we propose DEEM, a simple but effective approach that utilizes the generative feedback of diffusion models to align the semantic distributions of the image encoder. This addresses the drawbacks of previous methods that solely relied on image encoders like CLIP-ViT, thereby enhancing the model's resilience against out-of-distribution samples and reducing visual hallucinations. Importantly, this is achieved without requiring additional training modules and with fewer training parameters. We extensively evaluated DEEM on both our newly constructed RobustVQA benchmark and other well-known benchmarks, POPE and MMVP, for visual hallucination and perception. In particular, DEEM improves LMM's visual perception performance to a large extent (e.g., 4\% ↑ on RobustVQA, 6.5\% ↑ on MMVP and 12.8 \% ↑ on POPE ). Compared to the state-of-the-art interleaved content generation models, DEEM exhibits enhanced robustness and a superior capacity to alleviate model hallucinations while utilizing fewer trainable parameters, less pre-training data (10\%), and a smaller base model size. Extensive experiments demonstrate that DEEM enhances the performance of LMMs on various downstream tasks without inferior performance in the long term, including visual question answering, image captioning, and text-conditioned image synthesis.

Selective state space models (SSM), such as Mamba, have gained prominence for their effectiveness in modeling sequential data. Despite their outstanding empirical performance, a comprehensive theoretical understanding of deep selective SSM remains elusive, hindering their further development and adoption for applications that need high fidelity. In this paper, we investigate the dynamical properties of tokens in a pre-trained Mamba model. In particular, we derive the dynamical system governing the continuous-time limit of the Mamba model and characterize the asymptotic behavior of its solutions. In the one-dimensional case, we prove that only one of the following two scenarios happens: either all tokens converge to zero, or all tokens diverge to infinity. We provide criteria based on model parameters to determine when each scenario occurs. For the convergent scenario, we empirically verify that this scenario negatively impacts the model's performance. For the divergent scenario, we prove that different tokens will diverge to infinity at different rates, thereby contributing unequally to the updates during model training. Based on these investigations, we propose two refinements for the model: excluding the convergent scenario and reordering tokens based on their importance scores, both aimed at improving practical performance. Our experimental results validate these refinements, offering insights into enhancing Mamba's effectiveness in real-world applications.

Vision language models (VLMs) have seen growing adoption in recent years, but many still struggle with basic spatial reasoning errors. We hypothesize that this is due to VLMs adopting pre-trained vision backbones, specifically vision transformers (ViTs) trained with image-level supervision and minimal inductive biases. Such models may fail to encode the class contents at each position in the image, and our goal is to resolve this with a vision backbone that effectively captures both local and global image semantics. Our main insight is that we do not require new supervision to learn this capability – pre-trained models contain significant knowledge of local semantics that we can extract and use for scalable self-supervision. We propose a new efficient post-training stage for ViTs called locality alignment and a novel fine-tuning procedure called MaskEmbed that uses a masked reconstruction loss to learn semantic contributions for each image patch. We first evaluate locality alignment with a vision-only benchmark, finding that it improves a model’s performance at patch-level semantic segmentation, especially for strong backbones trained with image-caption pairs (e.g., CLIP and SigLIP). We then train a series of VLMs with and without locality alignment, and show that locality-aligned backbones improve performance across a range of benchmarks, particularly ones that involve spatial understanding (e.g., RefCOCO, OCID-Ref, TallyQA, VSR, AI2D). Overall, we demonstrate that we can efficiently learn local semantic extraction via a locality alignment stage, and that this procedure benefits VLM training recipes that use off-the-shelf vision backbones.

Reinforcement learning (RL) has seen significant success across various domains, but its adoption is often limited by the black-box nature of neural network policies, making them difficult to interpret. In contrast, symbolic policies allow representing decision-making strategies in a compact and interpretable way. However, learning symbolic policies directly within on-policy methods remains challenging. In this paper, we introduce SYMPOL, a novel method for SYMbolic tree-based on-POLicy RL. SYMPOL employs a tree-based model integrated with a policy gradient method, enabling the agent to learn and adapt its actions while maintaining a high level of interpretability. We evaluate SYMPOL on a set of benchmark RL tasks, demonstrating its superiority over alternative tree-based RL approaches in terms of performance and interpretability. Unlike existing methods, it enables gradient-based, end-to-end learning of interpretable, axis-aligned decision trees within standard on-policy RL algorithms. Therefore, SYMPOL can become the foundation for a new class of interpretable RL based on decision trees. Our implementation is available under: https://github.com/s-marton/sympol

The proliferation of deepfakes and AI-generated content has led to a surge in media forgeries and misinformation, necessitating robust detection systems. However, current datasets lack diversity across modalities, languages, and real-world scenarios. To address this gap, we present ILLUSION (Integration of Life-Like Unique Synthetic Identities and Objects from Neural Networks), a large-scale, multi-modal deepfake dataset comprising 1.3 million samples spanning audio-visual forgeries, 26 languages, challenging noisy environments, and various manipulation protocols. Generated using 28 state-of-the-art generative techniques, ILLUSION includes faceswaps, audio spoofing, synchronized audio-video manipulations, and synthetic media while ensuring a balanced representation of gender and skin tone for unbiased evaluation. Using Jaccard Index and UpSet plot analysis, we demonstrate ILLUSION’s distinctiveness and minimal overlap with existing datasets, emphasizing its novel generative coverage. We benchmarked image, audio, video, and multi-modal detection models, revealing key challenges such as performance degradation in multilingual and multi-modal contexts, vulnerability to real-world distortions, and limited generalization to zero-day attacks. By bridging synthetic and real-world complexities, ILLUSION provides a challenging yet essential platform for advancing deepfake detection research. The dataset is publicly available at https://www.iab-rubric.org/illusion-database.

Large language models (LLMs) have been increasingly employed for (interactive) decision-making, via the development of LLM-based autonomous agents. Despite their emerging successes, the performance of LLM agents in decision-making has not been fully investigated through quantitative metrics, especially in the multi-agent setting when they interact with each other, a typical scenario in real-world LLM-agent applications. To better understand the limits of LLM agents in these interactive environments, we propose to study their interactions in benchmark decision-making settings in online learning and game theory, through the performance metric of regret. We first empirically study the no-regret behaviors of LLMs in canonical non-stochastic online learning problems, as well as the emergence of equilibria when LLM agents interact through playing repeated games. We then provide some theoretical insights into the no-regret behaviors of LLM agents, under certain assumptions on the supervised pre-training and the rationality model of human decision-makers who generate the data. Notably, we also identify (simple) cases where advanced LLMs such as GPT-4 fail to be no-regret. To further promote the no-regret behaviors, we propose a novel unsupervised training loss of regret-loss, which, in contrast to the supervised pre-training loss, does not require the labels of (optimal) actions. Finally, we establish the statistical guarantee of generalization bound for regret-loss minimization, and more importantly, the optimization guarantee that minimizing such a loss may automatically lead to known no-regret learning algorithms, when single-layer self-attention models are used. Our further experiments demonstrate the effectiveness of our regret-loss, especially in addressing the above “regrettable” cases.

The rise of Large Language Models (LLMs) has revolutionized numerous domains, yet these models still exhibit weakness in understanding structured tabular data. Although the growing context window promises to accommodate a larger volume of table contents, it does not inherently improve the model's ability to understand the underlying structure and semantics of tabular data. To bridge the semantic gap between **T**ext and **T**able, we propose **T**n**T**, a table-language model that features multimodal table representations to empower LLMs to effectively and efficiently abstract structure-enriched semantics from tabular data. **T**n**T** also introduces a scalable and efficient training pipeline, featuring novel self-supervised tasks, to integrate abstract tabular knowledge into the language modality. Extensive experimental results on NL2SQL demonstrate a much better table understanding of **T**n**T**, which achieves up to **14.4** higher execution accuracy compared with traditional text-based table representations.

Recently, data-driven simulators based on graph neural networks have gained attention in modeling physical systems on unstructured meshes. However, they struggle with long-range dependencies in fluid flows, particularly in refined mesh regions. This challenge, known as the 'over-squashing' problem, hinders information propagation. While existing graph rewiring methods address this issue to some extent, they only consider graph topology, overlooking the underlying physical phenomena. We propose Physics-Informed Ollivier--Ricci Flow (PIORF), a novel rewiring method that combines physical correlations with graph topology. PIORF uses Ollivier--Ricci curvature (ORC) to identify bottleneck regions and connects these areas with nodes in high-velocity gradient nodes, enabling long-range interactions and mitigating over-squashing. Our approach is computationally efficient in rewiring edges and can scale to larger simulations. Experimental results on 3 fluid dynamics benchmark datasets show that PIORF consistently outperforms baseline models and existing rewiring methods, achieving up to 26.2\% improvement.

Offline reinforcement learning (RL) aims to learn from static datasets and thus faces the challenge of value estimation errors for out-of-distribution actions. The in-sample learning scheme addresses this issue by performing implicit TD backups that does not query the values of unseen actions. However, pre-existing in-sample value learning and policy extraction methods suffer from over-regularization, limiting their performance on suboptimal or compositional datasets. In this paper, we analyze key factors in in-sample learning that might potentially hinder the use of a milder constraint. We propose Actor-Critic with Temperature adjustment and In-sample Value Ensemble (ACTIVE), a novel in-sample offline RL algorithm that leverages an ensemble of $V$-functions for critic training and adaptively adjusts the constraint level using dual gradient descent. We theoretically show that the $V$-ensemble suppresses the accumulation of initial value errors, thereby mitigating overestimation. Our experiments on the D4RL benchmarks demonstrate that ACTIVE alleviates overfitting of value functions and outperforms existing in-sample methods in terms of learning stability and policy optimality.

Graph Neural Networks (GNNs) deliver strong classification results but often suffer from poor calibration performance, leading to overconfidence or underconfidence. This is particularly problematic in high-stakes applications where accurate uncertainty estimates are essential. Existing post-hoc methods, such as temperature scaling, fail to effectively utilize graph structures, while current GNN calibration methods often overlook the potential of leveraging diverse input information and model ensembles jointly. In the paper, we propose Graph Ensemble Temperature Scaling (GETS), a novel calibration framework that combines input and model ensemble strategies within a Graph Mixture-of-Experts (MoE) architecture. GETS integrates diverse inputs, including logits, node features, and degree embeddings, and adaptively selects the most relevant experts for each node’s calibration procedure. Our method outperforms state-of-the-art calibration techniques, reducing expected calibration error (ECE) by $\geq$ 25% across 10 GNN benchmark datasets. Additionally, GETS is computationally efficient, scalable, and capable of selecting effective input combinations for improved calibration performance. The implementation is available at https://github.com/ZhuangDingyi/GETS/.

We provide a framework for solving inverse problems with diffusion models learned from linearly corrupted data. Firstly, we extend the Ambient Diffusion framework to enable training directly from measurements corrupted in the Fourier domain. Subsequently, we train diffusion models for MRI with access only to Fourier subsampled multi-coil measurements at acceleration factors R$=2, 4, 6, 8$. Secondly, we propose $\textit{Ambient Diffusion Posterior Sampling}$ (A-DPS), a reconstruction algorithm that leverages generative models pre-trained on one type of corruption (e.g. image inpainting) to perform posterior sampling on measurements from a different forward process (e.g. image blurring). For MRI reconstruction in high acceleration regimes, we observe that A-DPS models trained on subsampled data are better suited to solving inverse problems than models trained on fully sampled data. We also test the efficacy of A-DPS on natural image datasets (CelebA, FFHQ, and AFHQ) and show that A-DPS can sometimes outperform models trained on clean data for several image restoration tasks in both speed and performance.

Causal inference from observational data has attracted considerable attention among researchers. One main obstacle is the handling of confounders. As direct measurement of confounders may not be feasible, recent methods seek to address the confounding bias via proxy variables, i.e., covariates postulated to be conducive to the inference of latent confounders. However, the selected proxies may scramble both confounders and post-treatment variables in practice, which risks biasing the estimation by controlling for variables affected by the treatment. In this paper, we systematically investigate the bias due to latent post-treatment variables, i.e., latent post-treatment bias, in causal effect estimation. Specifically, we first derive the bias when selected proxies scramble both latent confounders and post-treatment variables, which we demonstrate can be arbitrarily bad. We then propose a Confounder-identifiable VAE (CiVAE) to address the bias. Based on a mild assumption that the prior of latent variables that generate the proxy belongs to a general exponential family with at least one invertible sufficient statistic in the factorized part, CiVAE individually identifies latent confounders and latent post-treatment variables up to bijective transformations. We then prove that with individual identification, the intractable disentanglement problem of latent confounders and post-treatment variables can be transformed into a tractable independence test problem despite arbitrary dependence may exist among them. Finally, we prove that the true causal effects can be unbiasedly estimated with transformed confounders inferred by CiVAE. Experiments on both simulated and real-world datasets demonstrate significantly improved robustness of CiVAE.

In this paper, we introduce the Generalized Video Moment Retrieval (GVMR) framework, which extends traditional Video Moment Retrieval (VMR) to handle a wider range of query types. Unlike conventional VMR systems, which are often limited to simple, single-target queries, GVMR accommodates both non-target and multi-target queries. To support this expanded task, we present the NExT-VMR dataset, derived from the YFCC100M collection, featuring diverse query scenarios to enable more robust model evaluation. Additionally, we propose BCANet, a transformer-based model incorporating the novel Boundary-aware Cross Attention (BCA) module. The BCA module enhances boundary detection and uses cross-attention to achieve a comprehensive understanding of video content in relation to queries. BCANet accurately predicts temporal video segments based on natural language descriptions, outperforming traditional models in both accuracy and adaptability. Our results demonstrate the potential of the GVMR framework, the NExT-VMR dataset, and BCANet to advance VMR systems, setting a new standard for future multimedia information retrieval research.

The quality of generative models depends on the quality of the data they are trained on. Creating large-scale, high-quality datasets is often expensive and sometimes impossible, e.g.~in certain scientific applications where there is no access to clean data due to physical or instrumentation constraints. Ambient Diffusion and related frameworks train diffusion models with solely corrupted data (which are usually cheaper to acquire) but ambient models significantly underperform models trained on clean data. We study this phenomenon at scale by training more than $80$ models on data with different corruption levels across three datasets ranging from $30,000$ to $\approx 1.3$M samples. We show that it is impossible, at these sample sizes, to match the performance of models trained on clean data when only training on noisy data. Yet, a combination of a small set of clean data (e.g.~$10\%$ of the total dataset) and a large set of highly noisy data suffices to reach the performance of models trained solely on similar-size datasets of clean data, and in particular to achieve near state-of-the-art performance. We provide theoretical evidence for our findings by developing novel sample complexity bounds for learning from Gaussian Mixtures with heterogeneous variances. Our theoretical model suggests that, for large enough datasets, the effective marginal utility of a noisy sample is exponentially worse that of a clean sample. Providing a small set of clean samples can significantly reduce the sample size requirements for noisy data, as we also observe in our experiments.

Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its effects on the learned policy remain inconsistent: some studies find it crucial for achieving strong results, while others observe decreased performance. In this paper, we first dissect how action chunking impacts the divergence between a learner and a demonstrator. We find that action chunking allows the learner to better capture the temporal dependencies in demonstrations but at the cost of reduced reactivity to unexpected states. To address this tradeoff, we propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop adaptation. At each timestep, BID samples multiple candidate predictions and searches for the optimal one based on two criteria: (i) backward coherence, which favors samples that align with previous decisions; (ii) forward contrast, which seeks samples of high likelihood for future plans. By coupling decisions within and across action chunks, BID promotes both long-term consistency and short-term reactivity. Experimental results show that our method boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks. Code and videos are available at https://bid-robot.github.io.

Joint finetuning of a pretrained encoder and a randomly initialized decoder has been the de facto standard in semantic segmentation, but the vulnerability of this approach to domain shift has not been studied. We investigate the vulnerability issue of joint finetuning, and propose a novel finetuning framework called Decoupled FineTuning (DeFT) for domain generalization as a solution. DeFT operates in two stages. Its first stage warms up the decoder with the frozen, pretrained encoder so that the decoder learns task-relevant knowledge while the encoder preserves its generalizable features. In the second stage, it decouples finetuning of the encoder and decoder into two pathways, each of which concatenates an adaptive component (AC) and retentive component (RC); the encoder and decoder play different roles between AC and RC in different pathways. ACs are updated by gradients of the loss on the source domain, while RCs are updated by exponential moving average biased toward their initialization to retain their generalization capability. By the two separate optimization pathways with opposite AC-RC configurations, DeFT reduces the number of learnable parameters virtually, and decreases the distance between learned parameters and their initialization, leading to improved generalization capability. DeFT significantly outperformed existing methods in various domain shift scenarios, and its performance was further boosted by incorporating a simple distance regularization.

Large Language Models (LLMs) have demonstrated remarkable capabilities through pretraining and alignment. However, superior short-context LLMs may underperform in long-context scenarios due to insufficient long-context alignment. This alignment process remains challenging due to the impracticality of human annotation for extended contexts and the difficulty in balancing short- and long-context performance. To address these challenges, we introduce LongPO, that enables short-context LLMs to self-evolve to excel on long-context tasks by internally transferring short-context capabilities. LongPO harnesses LLMs to learn from self-generated short-to-long preference data, comprising paired responses generated for identical instructions with long-context inputs and their compressed short-context counterparts, respectively. This preference reveals capabilities and potentials of LLMs cultivated during short-context alignment that may be diminished in under-aligned long-context scenarios. Additionally, LongPO incorporates a short-to-long KL constraint to mitigate short-context performance decline during long-context alignment. When applied to Mistral-7B-Instruct-v0.2 from 128K to 512K context lengths, LongPO fully retains short-context performance and largely outperforms naive SFT and DPO in both long- and short-context tasks. Specifically, LongPO-trained models can achieve results on long-context benchmarks comparable to, or even surpassing, those of superior LLMs (e.g., GPT-4-128K) that involve extensive long-context annotation and larger parameter scales. Our code is available at https://github.com/DAMO-NLP-SG/LongPO.

The primary focus of most recent works on open-vocabulary neural fields is extracting precise semantic features from the VLMs and then consolidating them efficiently into a multi-view consistent 3D neural fields representation. However, most existing works over-trusted SAM to regularize image-level CLIP without any further refinement. Moreover, several existing works improved efficiency by dimensionality reduction of semantic features from 2D VLMs before fusing with 3DGS semantic fields, which inevitably leads to multi-view inconsistency. In this work, we propose econSG for open-vocabulary semantic segmentation with 3DGS. Our econSG consists of: 1) A Confidence-region Guided Regularization (CRR) that mutually refines SAM and CLIP to get the best of both worlds for precise semantic features with complete and precise boundaries. 2) A low dimensional contextual space to enforce 3D multi-view consistency while improving computational efficiency by fusing backprojected multi-view 2D features and follow by dimensional reduction directly on the fused 3D features instead of operating on each 2D view separately. Our econSG show state-of-the-art performance on four benchmark datasets compared to the existing methods. Furthermore, we are also the most efficient training among all the methods.