... International Conference on Learning Representations最新文献

{"title":"ProReGen: Progressive Residual Generation under Attribute Correlations.","authors":"Ruby Shrestha, Ajay Gopi, Casey Meisenzahl, Bipin Lekhak, Linwei Wang","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Attribute correlations in the training data will compromise the ability of a deep generative model (DGM) to synthesize images with under-represented attribute combinations (<i>i.e</i>., minority samples). Existing approaches mitigate this by data re-sampling to remove attribute correlations seen by the DGM, using a classifier to provide <i>pseudo-supervision</i> on generated counterfactual samples, or incorporating inductive bias to explicitly decompose the generation into independent submechanisms. We present ProReGen, a <i>progressive residual generation</i> approach inspired by the classical Robinson's transformation, to partial out from an image attribute <math> <msub><mrow><mi>x</mi></mrow> <mrow><mn>2</mn></mrow> </msub> </math> its component <math><mi>m</mi> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> </mrow> </mfenced> </math> that is predictable by other image attributes <math> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> </math> , and the residual <math><mi>γ</mi> <mo>=</mo> <msub><mrow><mi>x</mi></mrow> <mrow><mn>2</mn></mrow> </msub> <mo>-</mo> <mi>m</mi> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> </mrow> </mfenced> </math> that is not. This simplifies the problem of learning a DGM <math><mi>g</mi> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> <mo>,</mo> <msub><mrow><mi>x</mi></mrow> <mrow><mn>2</mn></mrow> </msub> </mrow> </mfenced> </math> conditioned on correlated inputs, to learning <math> <mover><mrow><mi>g</mi></mrow> <mo>~</mo></mover> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> <mo>,</mo> <mi>γ</mi></mrow> </mfenced> </math> conditioned on orthogonal inputs. It further allows us to progressively learn <math> <mover><mrow><mi>g</mi></mrow> <mo>~</mo></mover> </math> by first shifting the burden to abundant majority samples to learn <math> <mover><mrow><mi>g</mi></mrow> <mo>~</mo></mover> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> <mo>,</mo> <mi>γ</mi> <mo>=</mo> <mn>0</mn></mrow> </mfenced> </math> , and then expanding it with additional layers <math> <msub><mrow><mi>g</mi></mrow> <mrow><mtext>res</mtext></mrow> </msub> </math> to resolve its difference to <math> <mover><mrow><mi>g</mi></mrow> <mo>~</mo></mover> <mfenced> <mrow> <msub><mrow><mi>x</mi></mrow> <mrow><mn>1</mn></mrow> </msub> <mo>,</mo> <mi>γ</mi></mrow> </mfenced> </math> using residual attribute <math><mi>γ</mi></math> on limited minority samples. On three benchmark datasets with varying strengths of attribute correlation and one dataset with natural attribute correlation, we demonstrate that ProReGen-with input orthogonalization and progressive residual learning-improved the correctness of minority generations compared to existing strategies.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"6667-6695"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13440320/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148681903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Graph-of-Agents: A Graph-based Framework for Multi-Agent LLM Collaboration.","authors":"Sukwon Yun, Jie Peng, Pingzhi Li, Wendong Fan, Jie Chen, James Zou, Guohao Li, Tianlong Chen","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>With an ever-growing zoo of LLMs and benchmarks, the need to orchestrate multiple models for improved task performance has never been more pressing. While frameworks like Mixture-of-Agents (MoA) attempt to coordinate LLMs, they often fall short in terms of (1) selecting relevant agents, (2) facilitating effective intra-agent communication, and (3) integrating responses efficiently. In this work, we propose Graph-of-Agents (GoA), a new graph-based framework for modeling multi-agent LLM communication. Our approach begins with node sampling, selecting only the most relevant agents by leveraging model cards that summarize each model's domain, task specialization, and other characteristics. Next, we construct edges between the selected agents by evaluating their responses against one another to determine relevance ordering. Directed message passing is then performed from highly relevant agents to less relevant ones to enhance their responses, followed by reverse message passing to refine the original responses of the more relevant agents. Finally, the updated responses are aggregated via graph-based pooling (e.g., max or mean pooling) to produce a single, unified answer. We evaluate GoA on diverse multi-domain benchmarks (MMLU, MMLU-Pro, GPQA) and domain-specific benchmarks (MATH, HumanEval, MedMCQA), with an agent pool of 6 LLMs spanning multiple domains. Surprisingly, GoA achieves superior performance using only 3 selected agents, outperforming recent multi-agent LLM baselines that utilize all 6 agents simultaneously. By adopting a graph structure, GoA offers both scalability and effectiveness through structured message passing-positioning it as a strong candidate for navigating the challenges of the ever-growing LLM zoo. Code is available at: https://github.com/UNITES-Lab/GoA.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"19745-19760"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13459884/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning to Recall with Transformers Beyond Orthogonal Embeddings.","authors":"Nuri Mert Vural, Alberto Bietti, Mahdi Soltanolkotabi, Denny Wu","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Modern large language models (LLMs) excel at tasks that require storing and retrieving knowledge, such as factual recall and question answering. Transformers are central to this capability because they can encode information during training and retrieve it at inference. Existing theoretical analyses typically study transformers under idealized assumptions such as infinite data or orthogonal embeddings. In realistic settings, however, models are trained on finite datasets with non-orthogonal (random) embeddings. We address this gap by analyzing a single-layer transformer with random embeddings trained with (empirical) gradient descent on a simple token-retrieval task, where the model must identify an informative token within a length- <math><mi>L</mi></math> sequence and learn a one-to-one mapping from tokens to labels. Our analysis tracks the \"early phase\" of gradient descent and yields explicit formulas for the model's storage capacity-revealing a multiplicative dependence between sample size <math><mi>N</mi></math> , embedding dimension <math><mi>d</mi></math> , and sequence length <math><mi>L</mi></math> . We validate these scalings numerically and further complement them with a lower bound for the underlying statistical problem, demonstrating that this multiplicative scaling is intrinsic under non-orthogonal embeddings. Code to reproduce all experiments is publicly available.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"75057-75129"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460266/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"InputDSA : Demixing then comparing recurrent and externally driven dynamics.","authors":"Ann Huang, Mitchell Ostrow, Satpreet H Singh, Leo Kozachkov, Ila Fiete, Kanaka Rajan","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>In control problems and basic scientific modeling, it is important to compare observations with dynamical simulations. For example, comparing two neural systems can shed light on the nature of emergent computations in the brain and deep neural networks. Recently, Ostrow et al. (2023) introduced Dynamical Similarity Analysis (DSA), a method to measure the similarity of two systems based on their recurrent dynamics rather than geometry or topology. However, DSA does not consider how inputs affect the dynamics, meaning that two similar systems, if driven differently, may be classified as different. Because real-world dynamical systems are rarely autonomous, it is important to account for the effects of input drive. To this end, we introduce a novel metric for comparing both intrinsic (recurrent) and input-driven dynamics, called InputDSA (iDSA). InputDSA extends the DSA framework by estimating and comparing both input and intrinsic dynamic operators using a variant of Dynamic Mode Decomposition with control (DMDc) based on subspace identification. We demonstrate that InputDSA can successfully compare partially observed, input-driven systems from noisy data. We show that when the true inputs are unknown, surrogate inputs can be substituted without a major deterioration in similarity estimates. We apply InputDSA on Recurrent Neural Networks (RNNs) trained with Deep Reinforcement Learning, identifying that high-performing networks are dynamically similar to one another, while low-performing networks are more diverse. Lastly, we apply InputDSA to neural data recorded from rats performing a cognitive task, demonstrating that it identifies a transition from input-driven evidence accumulation to intrinsically-driven decision-making. Our work demonstrates that InputDSA is a robust and efficient method for comparing intrinsic dynamics and the effect of external input on dynamical systems.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"55010-55053"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460255/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TimeRecipe: A Time-Series Forecasting Recipe via Benchmarking Module Level Effectiveness.","authors":"Zhiyuan Zhao, Juntong Ni, Shangqing Xu, Haoxin Liu, Wei Jin, B Aditya Prakash","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Time-series forecasting is an essential task with wide real-world applications across domains. While recent advances in deep learning have enabled time-series forecasting models with accurate predictions, there remains considerable debate over which architectures and design components, such as series decomposition or normalization, are most effective under varying conditions. Existing benchmarks primarily evaluate models at a high level, offering limited insight into why certain designs work better. To mitigate this gap, we propose TimeRecipe, a unified benchmarking framework that systematically evaluates time-series forecasting methods at the module level. TimeRecipe conducts over 10,000 experiments to assess the effectiveness of individual components across a diverse range of datasets, forecasting horizons, and task settings. Our results reveal that exhaustive exploration of the design space can yield models that outperform existing state-of-the-art methods and uncover meaningful intuitions linking specific design choices to forecasting scenarios. Furthermore, we release a practical toolkit within TimeRecipe that recommends suitable model architectures based on these empirical insights.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"150745-150792"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13489589/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148805457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledgeable Language Models as Black-Box Optimizers for Personalized Medicine.","authors":"Michael S Yao, Osbert Bastani, Alma Andersson, Tommaso Biancalani, Aïcha Bentaieb, Claudia Iriondo","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>The goal of personalized medicine is to discover a treatment regimen that optimizes a patient's clinical outcome based on their personal genetic and environmental factors. However, candidate treatments cannot be arbitrarily administered to the patient to assess their efficacy; we often instead have access to an <i>in silico</i> surrogate model that approximates the true fitness of a proposed treatment. Unfortunately, such surrogate models have been shown to fail to generalize to previously unseen patient-treatment combinations. We hypothesize that domain-specific prior knowledge-such as medical textbooks and biomedical knowledge graphs-can provide a meaningful alternative signal of the fitness of proposed treatments. To this end, we introduce <b>L</b>LM-based <b>E</b>ntropy-guided <b>O</b>ptimization with k<b>N</b>owledgeable priors (<b>LEON</b>), a mathematically principled approach to leverage large language models (LLMs) as black-box optimizers without any task-specific fine-tuning, taking advantage of their ability to contextualize unstructured domain knowledge to propose personalized treatment plans in natural language. In practice, we implement LEON via 'optimization by prompting,' which uses LLMs as stochastic engines for proposing treatment designs. Experiments on real-world optimization tasks show LEON outperforms both traditional and LLM-based methods in proposing individualized treatments for patients.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"50440-50502"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13461248/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Controllable diffusion-based generation for multi-channel biological data.","authors":"Haoran Zhang, Mingyuan Zhou, Wesley Tansey","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Biological profiling technologies, such as imaging mass cytometry (IMC) and spatial transcriptomics (ST), generate multi-channel data with strong spatial alignment and complex inter-channel relationships. Modeling such data requires generative frameworks that jointly model spatial structure and inter-channel dependencies and generalize across arbitrary subsets of observed and missing channels. Existing generative models typically assume low-dimensional inputs (e.g., RGB images) and rely on simple conditioning mechanisms that disrupt spatial correspondence and overlook inter-channel dependencies. This work proposes a unified multi-channel diffusion (MCD) framework for controllable generation of structured biological data with complex inter-channel relationships. Our model introduces two key innovations: (1) a hierarchical feature injection mechanism that enables multi-resolution conditioning on spatially aligned observed channels, and (2) two complementary channel attention modules to capture inter-channel relationships and recalibrate latent features. To support flexible conditioning and generalization to arbitrary sets of observed channels, we train the model using a random channel masking strategy, enabling it to reconstruct missing channels given any combination of observed channels as the spatial condition. We demonstrate state-of-the-art performance across both spatial and non-spatial biological data generation tasks, including imputation in spatial proteomics and clinical imaging, as well as gene-to-protein translation in single-cell datasets, and show strong generalizability to unseen conditional configurations.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"149825-149845"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13459956/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Shortcut Diffusion Training With Cumulative Consistency Loss: An Optimal Control View.","authors":"Paribesh Regmi, Sandesh Ghimire, Rui Li","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Although iterative denoising (i.e., diffusion/flow) methods offer strong generative performance, they suffer from low generation efficiency, requiring hundreds of steps of network forward passes to simulate a single sample. Mitigating this requires taking larger step-sizes during simulation, thereby allowing one- or few-step generation. Recently proposed shortcut model learns larger step-sizes by enforcing alignment between its direction and the path defined by a base many-step flow-matching model through a self-consistency loss. However, its generation quality is significantly lower than the base model. In this paper, we formulate few-step generation as a controlled base generative process, and show that self-consistency loss can be understood through the lens of optimal control. This perspective naturally motivates its generalization to the proposed cumulative self-consistency loss that cumulatively penalizes misalignment along the entire trajectory. This encourages larger step-sizes that not only align with the base model at the current time step but also support alignment in the subsequent steps, facilitating high-quality generation. Furthermore, we draw a connection between our approach and reinforcement learning, potentially opening the door to a new set of approaches for few-step generation. Experiments show that we significantly improve one- and few-step generation quality under the same training budget. Implementation is available at: https://github.com/paribeshregmi/Shortcut-CSL.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"68425-68444"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13461247/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CellDuality: Unlocking Biological Reasoning in LLMs with Self-Supervised RLVR.","authors":"Yuhang Chen, Zhen Tan, Ruichen Zhang, Mufan Qiu, Tianlong Chen","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Developing generalist large language models (LLMs) capable of complex biological reasoning is a central challenge in computational biology. While existing LLMs excel at predictive tasks like cell type annotation and logically-constrained problems, enabling open-ended and mechanistic reasoning remains a challenge. A promising direction is Reinforcement Learning from Verifiable Rewards (RLVR), which has been shown to significantly enhance complex reasoning in general domains like mathematics and code synthesis. However, its application in biology is hindered, as most biological outcomes are non-verifiable. For example, verifying a generated gene sequence is usually infeasible. In this paper, we introduce CellDuality, a self-supervised framework that enables LLM agents for robust reasoning in single-cell biology. Our framework is built on the principle of complementary task duality, a self-verification process that leverages a bidirectional reasoning loop. First, the model performs a forward reasoning task by predicting a biological outcome (e.g., a cell's response to a drug). Then, in a complementary inverse task, it must reason backward from its own prediction to reconstruct the initial conditions (e.g., the original drug perturbation). The fidelity of this reconstruction serves as an intrinsic reward signal, creating a feedback loop that enforces logical and biological consistency. We use these intrinsic rewards to align the base LLM via reinforcement learning, without requiring ground-truth verification labels. We demonstrate that CellDuality achieves state-of-the-art performance and provides coherent biological explanations across a diverse suite of single-cell reasoning tasks. Critically, on the challenging out-of-distribution perturbation prediction benchmark, our self-supervised approach significantly outperforms the standard fine-tuning baseline and narrows the performance gap to a supervised RLVR baseline. Our work showcases a new path toward scalable training of biological foundation models.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"115272-115283"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13441433/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148681663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generating metamers of human scene understanding.","authors":"Ritik Raina, Abe Leite, Alexandros Graikos, Seoyoung Ahn, Dimitris Samaras, Gregory J Zelinsky","doi":"","DOIUrl":"","url":null,"abstract":"<p><p>Human vision combines low-resolution \"gist\" information from the visual periphery with sparse but high-resolution information from fixated locations to construct a coherent understanding of a visual scene. In this paper, we introduce <i>MetamerGen</i>, a tool for generating scenes that are aligned with latent human scene representations. <b><i>MetamerGen</i></b> is a latent diffusion model that combines peripherally obtained scene gist information with information obtained from scene-viewing fixations to generate image metamers for what humans understand after viewing a scene. Generating images from both high and low resolution (i.e. \"foveated\") inputs constitutes a novel image-to-image synthesis problem, which we tackle by introducing a dual-stream representation of the foveated scenes consisting of DINOv2 tokens that fuse detailed features from fixated areas with peripherally degraded features capturing scene context. To evaluate the perceptual alignment of <b><i>MetamerGen</i></b> generated images to latent human scene representations, we conducted a same-different behavioral experiment where participants were asked for a \"same\" or \"different\" response between the generated and the original image. With that, we identify scene generations that are indeed <i>metamers</i> for the latent scene representations formed by the viewers. <b><i>MetamerGen</i></b> is a powerful tool for understanding scene understanding. Our proof-of-concept analyses uncovered specific features at multiple levels of visual processing that contributed to human judgments. While it can generate metamers even conditioned on random fixations, we find that high-level semantic alignment most strongly predicts metamerism when the generated scenes are conditioned on viewers' own fixated regions.</p>","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"154675-154700"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460306/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}