ICLR 2026 Papers

Open-Vocabulary Object Detection (OVOD) plays a critical role in autonomous driving and human-computer interaction by enabling perception beyond closed-set categories. However, current approaches predominantly rely on multimodal fusion, facing dual limitations: multimodal fusion methods incur heavy computational overhead from text encoders, while task-coupled designs compromise between detection precision and open-world generalization. To address these challenges, we propose Decoupled Cognition DETR, a vision framework featuring a three-stage cognitive distillation mechanism: Dynamic Hierarchical Concept Pool constructs self-evolving concept prototypes using LLaVA-generated region descriptions filtered by CLIP alignment, aiming to replace costly text encoders and reduce computational overhead; Hierarchical Knowledge Distillation decouples visual-semantic space mapping via prototype-centric projection, avoiding task coupling to enhance open-world generalization; Parametric Decoupling Training coordinates localization and cognition through dual-stream gradient isolation, further optimizing detection precision. Extensive experiments on the common OVOD evaluation protocol demonstrated that DeCo-DETR achieves state-of-the-art performance compared to existing OVOD methods. It provides a new paradigm for extending OVOD to more real-world applications.

Gradient descent (GD) is a fundamental optimization method in deep learning, yet its asymptotic directional properties remain less understood. In this paper, we prove that if GD converges, its trajectory either aligns toward a fixed direction or oscillates along a specific line. The fixed-direction convergence occurs under small learning rates, while the oscillatory convergence behavior emerges for large learning rates. This result offers a new lens for understanding long-term GD dynamics. Experimentally, we find that this directional convergence behavior also appears in stochastic gradient descent (SGD) and Adam. Furthermore, we discuss how these theoretical findings regarding oscillatory convergence might offer a perspective on the sharpness dynamics observed in the Edge of Stability (EoS) regime. Our work provides both theoretical clarity and practical insight into the behavior of dynamics for multiple optimization methods.

Reward models trained on human preference data have demonstrated strong effectiveness in aligning Large Language Models (LLMs) with human intent under the framework of Reinforcement Learning from Human Feedback (RLHF). However, RLHF remains vulnerable to reward hacking, where the policy exploits imperfections in the reward function rather than genuinely learning the intended behavior. Although significant efforts have been made to mitigate reward hacking, they predominantly focus on and evaluate in-distribution scenarios, where the training and testing data for the reward model share the same distribution. In this paper, we empirically show that state-of-the-art methods struggle in more challenging out-of-distribution (OOD) settings. We further demonstrate that incorporating fine-grained multi-attribute scores helps address this challenge. However, the limited availability of high-quality data often leads to weak performance of multi-objective reward functions, which can negatively impact overall performance and become the bottleneck. To address this issue, we propose a unified reward modeling framework that jointly trains Bradley-Terry (BT) single-objective and multi-objective regression-based reward functions using a shared embedding space. We theoretically establish a connection between the BT loss and the regression objective and highlight their complementary benefits. Specifically, the regression task enhances the single-objective reward function’s ability to mitigate reward hacking in challenging OOD settings, while BT-based training improves the scoring capability of the multi-objective reward function, enabling a 7B model to outperform a 70B baseline. Extensive experimental results demonstrate that our framework significantly improves both the robustness and the scoring performance of reward models.

Linear sequence modeling methods, such as linear attention, state space modeling, and linear RNNs, offer significant efficiency improvements by reducing the complexity of training and inference. However, these methods typically compress the entire input sequence into a single fixed-size memory state, which leads to suboptimal performance on recall-intensive tasks. To address this limitation, we introduce a novel architecture called Mixture-of-Memories (MoM). MoM utilizes multiple independent memory states, with a router network directing input tokens to specific memory states. This approach greatly enhances the overall memory capacity while minimizing memory interference. MoM serves as a general framework that can be seamlessly combined with diverse memory update mechanisms across linear models. As a result, MoM performs exceptionally well on recall-intensive tasks, surpassing existing linear sequence modeling techniques. Despite incorporating multiple memory states, the computation of each memory state remains linear in complexity, allowing MoM to retain the linear-complexity advantage during training, while constant-complexity during inference. Our experimental results show that MoM outperforms current linear sequence models on downstream language tasks, particularly recall-intensive tasks, and even achieves performance comparable to Transformer models.

Reinforcement learning (RL) has emerged as a central paradigm for training large language models (LLMs) in reasoning tasks. Yet recent studies question RL’s ability to incentivize reasoning capacity beyond the base model. This raises a key challenge: how can RL be adapted to solve harder reasoning problems more effectively? To address this challenge, we propose a simple yet effective strategy via Question Augmentation: introduce partial solutions during training to reduce problem difficulty and provide more informative learning signals. Our method, QuestA, when applied during RL training on math reasoning tasks, not only improves pass@1 but also pass@k—particularly on problems where standard RL struggles to make progress. This enables continual improvement over strong open-source models such as DeepScaleR and OpenMath Nemotron, further enhancing their reasoning capabilities. We achieve new state-of-the-art results on math benchmarks using 1.5B-parameter models: 72.50\% (+10.73\%) on AIME24, 62.29\% (+12.79\%) on AIME25, and 41.67\% (+10.11\%) on HMMT25. Code, data and model are available at https://anonymous.4open.science/r/questa932.

Referring Video Object Segmentation (RVOS) requires segmenting specific objects in a video guided by a natural language description. The core challenge of RVOS is to anchor abstract linguistic concepts onto a specific set of pixels and continuously segment them through the complex dynamics of a video. Faced with this difficulty, prior work has often decomposed the task into a pragmatic `locate-then-segment' pipeline. However, this cascaded design creates an information bottleneck by simplifying semantics into coarse geometric prompts (e.g, point), and struggles to maintain temporal consistency as the segmenting process is often decoupled from the initial language grounding. To overcome these fundamental limitations, we propose FlowRVS, a novel framework that reconceptualizes RVOS as a conditional continuous flow problem. This allows us to harness the inherent strengths of pretrained T2V models, fine-grained pixel control, text-video semantic alignment, and temporal coherence. Instead of conventional generating from noise to mask or directly predicting mask, we reformulate the task by learning a direct, language-guided deformation from a video's holistic representation to its target mask. Our one-stage, generative approach achieves new state-of-the-art results across all major RVOS benchmarks. Specifically, achieving a J&F of 51.1 in MeViS (+1.6 over prior SOTA) and 73.3 in the zero shot Ref-DAVIS17 (+2.7), demonstrating the significant potential of modeling video understanding tasks as continuous deformation processes.

Reinforcement learning with verifiable rewards (RLVR) has proven effective in training large reasoning models (LRMs) by leveraging answer-verifiable signals to guide policy optimization, which, however, suffers from high annotation costs. To alleviate this problem, recent work has explored unsupervised RLVR methods that derive rewards solely from the model’s internal consistency, such as through entropy and majority voting. While seemingly promising, these methods often suffer from model collapse in the later stages of training, which may arise from the reinforcement of incorrect reasoning patterns in the absence of external supervision. In this work, we investigate a novel semi-supervised RLVR paradigm that utilizes a small labeled set to **guide** RLVR training on unlabeled samples. Our key insight is that supervised rewards are essential for stabilizing consistency-based training on unlabeled samples, ensuring that only reasoning patterns verified on labeled instances are incorporated into RL training. Technically, we propose an effective policy optimization algorithm TraPO that filters out reliable unlabeled samples by matching their learning trajectory similarity to labeled ones. Building on this, TraPO achieves remarkable data efficiency and strong generalization on nine advanced benchmarks. With only 1K labeled and 3K unlabeled samples, TraPO reaches 42.6% average accuracy, surpassing the best unsupervised method trained on 45K unlabeled samples (38.3%). Notably, when using 4K labeled and 12K unlabeled samples, TraPO even **outperforms the fully supervised model** trained on the full 45K labeled samples on all benchmarks, while using only **10%** of the labeled data.

Spiking Neural Networks (SNNs) are regarded as energy-efficient neural architectures due to their event-driven, spike-based computation paradigm. However, existing SNNs suffer from two fundamental limitations: (1) the constrained representational space imposed by binary spike firing mechanisms, which restricts the network's capacity to encode complex spatiotemporal patterns, and (2) the ineffective design of surrogate gradient functions that leads to gradient mismatch issues and suboptimal learning dynamics. To address these challenges, we propose the Parametric Invertible Transformation (PIT), which operates in a conjugate manner with neuronal dynamics to achieve adaptive modulation and augmented spike representations simultaneously. Second, we design an auxiliary gradient correction term to mitigate the gradient mismatch issue and oscillation phenomena during training. Moreover, we introduce a theoretical framework for analyzing the spatiotemporal representation space of SNNs. Extensive experiments on both static and neuromorphic datasets demonstrate state-of-the-art performance with our proposed method. This approach lays the theoretical foundation for expanding the spatiotemporal representations of SNNs, offering a viable pathway for developing low-latency and high-performance neuromorphic processing systems in resource-constrained environments.

As an important branch of unsupervised learning, deep clustering has seen substantial progress in recent years. However, the majority of current deep clustering methods operate under the assumption of balanced or near-balanced cluster distributions. This assumption contradicts the common long-tailed class distributions in real-world data, leading to severe performance degradation in deep clustering. Although many long-tailed learning methods have been proposed, these approaches typically rely on label information to differentiate treatment across different classes, which renders them inapplicable to deep clustering scenarios. How to re-weight the training of deep clustering models in an unsupervised setting remains an open challenge. To address this, we propose a mini-cluster guided long-tailed deep clustering method, termed MiniClustering. We introduce a specialized clustering head that divide data into much more clusters than the target number of clusters. These predicted clusters are referred to as mini-clusters. The mini-cluster-level predictions serve as the guide for estimating the appropriate weights for classes with varying degrees of long-tailedness. The weights are then incorporated to re-weight the self-training loss in model training. In this way, we can mitigate model bias by re-weighting gradients from different classes. We evaluate our method on multiple benchmark datasets with different imbalance ratios to demonstrate its effectiveness. Further, our method can be readily applied to the downstream of existing unsupervised representation learning frameworks for long-tailed deep clustering. It can also adapt label-dependent long-tailed learning methods to unsupervised clustering tasks by leveraging the estimated weights. *The code is available in the Supplementary Material.*

We study the problem of contextual online bilateral trade. At each round, the learner faces a seller-buyer pair and must propose a trade price without observing their private valuations for the item being sold. The goal of the learner is to post prices to facilitate trades between the two parties. Before posting a price, the learner observes a $d$-dimensional context vector that influences the agent's valuations. Prior work in the contextual setting has focused on linear valuation models, where valuations are linear functions of the context. We provide the first characterisation of a general nonparametric setting in which the buyer’s and seller’s valuations behave according to arbitrary Lipschitz functions of the context. We design an algorithm that leverages contextual information through a hierarchical tree construction and guarantees regret $\widetilde{O}(T^{(d-1)/d})$. Remarkably, our algorithm operates under two stringent features of the setting: (1) one-bit feedback, where the learner only observes whether a trade occurred or not, and (2) strong budget balance, where the learner cannot subsidize or profit from the market participants. We further provide a matching lower bound in the full-feedback setting, demonstrating the tightness of our regret bound.

The escalating scale and cost of Large Language Models (LLMs) training necessitate accurate pre-training prediction of downstream task performance for efficient resource allocation. This is challenged by: 1) the emergence phenomenon, where unpredictable capabilities appearing suddenly at critical model scales; and 2) uneven task difficulty and inconsistent performance scaling patterns, leading to high metric variability. Current prediction methods lack accuracy and reliability. We propose a Clustering-On-Difficulty (COD) framework for downstream performance prediction. The COD framework clusters tasks by their difficulty scaling features, thereby establishing a more stable and predictable subset that perform well-behaved scaling performance as compute increase. We adopt a performance scaling law to predict cluster-wise performance with theoretical support. Predictable subset performance acts as an intermediate predictor for the full evaluation set. We further derive a mapping function to accurately extrapolate the performance of the subset to the full set. Applied to an LLM with 70B parameters, COD achieved a 1.63\% average prediction error across eight key LLM benchmarks, offering actionable insights for resource allocation and training monitoring of LLMs pretraining.

Test-time scaling aims to improve language model performance by leveraging additional compute during inference. Many works have empirically studied techniques such as Best-of-N (BoN) and Rejection Sampling (RS) that make use of a verifier to enable test-time scaling. However, to date there is little theoretical understanding of how verifier *imperfection* affects performance — a gap we address in this work. Specifically, we prove that the instance-level accuracy of these methods is precisely characterized by the geometry of the verifier’s ROC curve. Our theory has two important takeaways, confirmed by experiments with Qwen and LLama models on GSM8K and MATH500. First, RS outperforms BoN for fixed compute, while both methods converge to the same accuracy in the infinite-compute limit. Second, it is generally impossible to predict the high-compute performance of either method based on observations in the low-compute regime.

Early action prediction seeks to anticipate an action before it fully unfolds, but limited visual evidence makes this task especially challenging. We introduce EAST, a simple and efficient framework that enables a model to reason about incomplete observations. In our empirical study, we identify key components when training early action prediction models. Our key contribution is a randomized training strategy that samples a time step separating observed and unobserved video frames, enabling a single model to generalize seamlessly across all test-time observation ratios. We further show that joint learning on both observed and future (oracle) representations significantly boosts performance, even allowing an encoder-only model to excel. To improve scalability, we propose a token masking procedure that cuts memory usage in half and accelerates training by 2× with no accuracy loss. Combined with a forecasting decoder, EAST sets a new state of the art on NTU60, SSv2, and UCF101, surpassing previous best work by 10.1, 7.7, and 3.9 percentage points, respectively. We support future research by releasing efficient training implementations and pre-trained models.

Recent progress in large reasoning models for challenging mathematical reasoning has been driven by reinforcement learning (RL). Incorporating long chain-of-thought (CoT) data during mid-training has also been shown to substantially improve reasoning depth. However, current approaches often utilize CoT data indiscriminately, leaving open the critical question of which data types most effectively enhance model reasoning capabilities. In this paper, we define the foundation model's reasoning potential for the first time as the inverse of the number of independent attempts required to correctly answer the question, which is strongly correlated with the final model performance. We then propose utilizing diverse data enriched with high-value reasoning patterns to expand the reasoning potential. Specifically, we abstract atomic reasoning patterns from CoT sequences, characterized by commonality and inductive capabilities, and use them to construct a core reference set enriched with valuable reasoning patterns. Furthermore, we propose a dual-granularity algorithm involving chains of reasoning patterns and token entropy, efficiently selecting high-value CoT data (CoTP) from the data pool that aligns with the core set, thereby training models to master reasoning effectively. Only 10B-token CoTP data enables the 85A6B Mixture-of-Experts (MoE) model to improve by **9.58\%** on the challenging AIME 2024 and 2025, and to raise the upper bound of downstream RL performance by **7.81\%**.

Deploying ML systems with only a few real samples makes operational metrics (such as alert rates or mean scores) highly unstable. Existing uncertainty quantification (UQ) methods fail here: frequentist intervals ignore the deployed predictive rule, Bayesian posteriors assume continual refitting, and conformal methods offer per-example rather than long-run guarantees. We introduce a forecast-first UQ framework that blends the empirical distribution with a frozen pretrained generator using a unique Dirichlet schedule, ensuring time-consistent forecasts. Uncertainty is quantified via martingale posteriors: a lightweight, likelihood-free resampling method that simulates future forecasts under the deployed rule, yielding sharp, well-calibrated intervals for both current and long-run metrics without retraining or density evaluation. A single hyperparameter, set by a small-$n$ minimax criterion, balances sampling variance and model--data mismatch; for bounded scores, we provide finite-time drift guarantees. We also show how this framework informs optimal retraining decisions. Applicable off-the-shelf to frozen generators (flows, diffusion, autoregressive models, GANs) and linear metrics (means, tails, NLL), it outperforms bootstrap baselines across vision and language benchmarks (WikiText-2, CIFAR-10, and SVHN datasets); e.g., it achieves $\sim$90\% coverage on GPT-2 with 20 samples vs.\ 37\% for bootstrap. Importantly, our uncertainty estimates are operational under the deployed forecasting rule agnostic of the population parameters, affording practicable estimators for deployment in real world settings.

Despite the rapid growth of machine learning research, corresponding code implementations are often unavailable, making it slow and labor-intensive for researchers to reproduce results and build upon prior work. In the meantime, recent Large Language Models (LLMs) excel at understanding scientific documents and generating high-quality code. Inspired by this, we introduce PaperCoder, a multi-agent LLM framework that transforms machine learning papers into operational code repositories. PaperCoder operates in three stages: planning, where it constructs a high-level roadmap, designs the system architecture with diagrams, identifies file dependencies, and generates configuration files; analysis, which focuses on interpreting implementation-specific details; and generation, where modular, dependency-aware code is produced. Moreover, each phase is instantiated through a set of specialized agents designed to collaborate effectively across the pipeline. We then evaluate PaperCoder on generating code implementations from machine learning papers based on both model-based and human evaluations, particularly from the authors of those papers, with author-released repositories as ground truth if available. Our results demonstrate the effectiveness of PaperCoder in creating high-quality, faithful implementations. Furthermore, it consistently shows strengths in the recently released PaperBench benchmark, surpassing strong baselines by substantial margins.

Image clustering is a fundamental task in visual machine learning. A key research direction in this field is the incorporation of prior knowledge. Recently, such prior knowledge has evolved from internal compactness constraints to external textual guidance. In particular, the introduction of textual modalities through CLIP has demonstrated impressive performance. However, CLIP is designed primarily for image–text alignment and may not be sufficient to capture clustering structures. Moreover, existing approaches often assume that textual features are universally beneficial, overlooking their varying suitability for different datasets. To address these issues, we propose to use spatial structure and selective text to jointly facilitate image clustering (SATC). Specifically, we design a graph attention network (GAT)-based encoder to capture relational dependencies among image patches, thereby extracting spatial features to facilitate clustering. In addition, we introduce a textual feature selector that uses the potential clustering compactness of textual features as the selection criterion and adaptively integrates them into the clustering process. Theoretical guidance is provided for this selector. Finally, the cluster assignment is produced through Tri-modal mutual distillation. Extensive experiments on 18 benchmark datasets demonstrate the effectiveness of SATC. The experimental results further verify the rationality of the textual feature selector. The code will be published.

Recent advances in Large Language Models (LLMs) and Reinforcement Learning (RL) have led to strong performance in open-domain question answering (QA). However, existing models still struggle with questions that admit multiple valid answers. Standard QA benchmarks, which typically assume a single gold answer, overlook this reality and thus produce inappropriate training signals. Existing attempts to handle ambiguity often rely on costly manual annotation, which is difficult to scale to multi-hop datasets such as HotpotQA and MuSiQue. In this paper, we present A$^2$Search, an annotation-free, end-to-end training framework to recognize and handle ambiguity. At its core is an automated pipeline that detects ambiguous questions and gathers alternative answers via trajectory sampling and evidence verification. The model is then optimized with RL using a carefully designed $\mathrm{AnsF1}$ reward, which naturally accommodates multiple answers. Experiments on eight open-domain QA benchmarks demonstrate that A$^2$Search achieves new state-of-the-art performance. With only a single rollout, A$^2$Search-7B yields an average $\mathrm{AnsF1}@1$ score of $48.4$% across four multi-hop benchmarks, outperforming all strong baselines, including the substantially larger ReSearch-32B ($46.2$%). Extensive analyses further show that A$^2$Search resolves ambiguity and generalizes across benchmarks, highlighting that embracing ambiguity is essential for building more reliable QA systems.

Large Language Models (LLMs) often experience compounding errors during long text generation. Early mistakes can propagate and lead to drift, faulty reasoning, or repetition. While scaling up models improves capabilities, it requires substantial computational resources, and the resulting self-correction behaviour remains unpredictable at inference time. Self-correction is a promising technique for addressing this issue. However, existing approaches have limitations. Supervised training methods can build self-correcting behaviours into models, but require training data collection and lack cross-domain generalizability. Current post-hoc iterative refinement methods operate only at inference time, but must wait for substantial portions of the draft to be generated before providing feedback. This feedback does not guarantee effective guidance, and the same mistake patterns can still reappear. In this paper, we introduce Once-More, a model-agnostic post-hoc self-correction framework that intervenes during generation. Once-More leverages token-level perplexity and feedback from verifiers to provide continuous guided steering of the generation path through a logit redistribution mechanism. This approach essentially helps accumulate "more correct" steps throughout the generation process. Evaluation on multiple benchmarks demonstrates that Once-More achieves state-of-the-art results compared to other self-correction methods. To our knowledge, Once-More is the first post-hoc method to leverage token perplexity and external feedback to perform continuous guided self-correction.