NeurIPS 2026 论文列表

Brain-to-speech decoding aims to restore communication by translating neural activity into speech-related representations. From a modeling perspective, this task can be formulated as a weakly supervised neural sequence decoding problem, where phoneme sequences must be recovered without frame-level annotations. Existing speech decoders commonly rely on recurrent sequence models to integrate temporal context, but their dense sequential computation can limit their suitability for resource-constrained portable or implantable neuroprosthetic systems. In this study, we propose a residual spiking neural network (SNN) for phoneme-level decoding of intracortical speech signals under constrained model capacity and weak sequence-level supervision. Our approach incorporates biologically inspired fixed connection-specific delays to integrate temporal context, providing each postsynaptic unit with access to recent presynaptic spike activity without recurrent state or dense temporal convolutions. Because each directed connection is assigned a single fixed delay, the temporal field is expanded without adding trainable delay parameters. Across 500k, 2M, and 5M core-parameter budgets, fixed delays consistently improve phoneme error rate (PER), with gains saturating once the delay range covers the task-relevant temporal horizon. Boundary-aligned CTC analysis further suggests that delayed temporal context reduces local uncertainty around phoneme transitions, indicating that fixed delays support weakly supervised phoneme alignment rather than simply increasing model capacity. Compared with recurrent, convolutional, and feedforward baselines under matched core-parameter budgets, the proposed SNN achieves the best PER at 500k and 2M parameters and the second-best PER at 5M parameters, while reducing estimated operation-level energy by approximately 2-4$\times$ relative to the strongest non-spiking baselines. These results suggest that fixed-delay SNNs provide an efficient temporal modeling mechanism for brain-to-speech decoding, potentially enabling lightweight decoding for on-device speech neuroprostheses.

Efficient benchmarking techniques aim to lower the computational cost of evaluating LLMs by predicting full benchmark scores using only a subset of a benchmark's questions. By reframing this problem as an instance of multiple regression with feature selection, we find that existing efficient benchmarking methods can be greatly improved by simply using kernel ridge regression at the prediction stage. Additionally, using an information-theoretic feature-selection algorithm called minimum redundancy maximum relevance (mRMR), we can further improve upon these methods by selecting question subsets that will be maximally useful for prediction. Except in very data-poor settings, these approaches consistently achieve smaller prediction errors (in both MAE and RMSE), and greater ranking correlation between predicted and true scores (in both Spearman $\rho$ and Kendall $\tau$) across a range of benchmarks using both binary and continuous metrics. Furthermore, mRMR subsampling is much faster than competitor methods (which often involve fitting probabilistic models or running clustering algorithms), and is more likely to select the same questions under different random seeds or training data splits.

The distributed setting for Saddle Problems (SPs) has recently emerged as a framework for modern applications in machine learning and multiagent systems. Despite its relevance, the theoretical foundations of this setting have not yet been thoroughly established. In this paper, we advance this research direction by formalizing the distributed setup for SPs and providing rigorous definitions of communication and oracle costs. Further, we prove lower bounds for any distributed gradient-span algorithm, which reveals the gap from existing methods and this theoretical limit. To this end, we provide a Decoupled Method built upon a novel multi-stage reduction that reduces the SP into a sequence of decoupled minimization tasks of residual norms. Our algorithm matches the communication lower bound, thus setting the communication complexity within the gradient-span algorithms. Moreover, it yields the first strict improvement over the long-standing oracle cost of the Extragradient method for general SPs. Finally, we study the extension of distributed SP into Variational Inequality Problem (VIP), which generalizes two-player zero-sum games to multiplayer general-sum games. We show that our Decoupled Method achieves a new state-of-the-art communication complexity for this broader class.

We consider the problem of maintaining a minimum-cost bipartite matching between two point sets $A$ and $B$ in a metric space $(\mathbb{X},\mathsf{d})$ under insertions and deletions of input points. The Wasserstein-$p$ ($W_p$) cost of a matching $M$ is defined as $(\sum_{(a,b)\in M}\mathsf{d}(a,b)^p)^{1/p}$ and an $\alpha$-approximate matching is one whose total cost is at most $\alpha$ times the cost of a minimum-cost matching. We obtain two main results. If $A$ and $B$ are points in $\mathbb{R}^d$ and the distance between two points is measured under the $\ell_p$-metric, an $O(d\epsilon^{-3/2})$-approximate $W_1$-matching between $A$ and $B$ can be maintained with an amortized $O\big(n^\epsilon d\epsilon^{-3/2}\varphi(n,1/\sqrt{\epsilon})\big)$ update time, for any parameter $\epsilon\in(0,1]$, and where $\varphi(n,1/\sqrt{\epsilon})$ denotes the update time of a $\epsilon^{-1/2}$-approximate nearest neighbor data structure. For the $\ell_2$ norm this results in an update time of $O\big((d+ \epsilon^{-3/2}n^\epsilon)\log n\big)$. The current state-of-the-art for high-dimensional settings only focus on the $\ell_2$-metric and maintain an $O(\log n)$-approximate matching. Our result not only extends to any $\ell_p$ norm, but it also improves the approximation factor for $d=o(\log n)$ under the $\ell_2$-metric. Next, we consider the problem of maintaining a $W_p$-matching when $A$ and $B$ are point sets in an arbitrary finite metric space. We show that for any $k,p\geq 1$ and constant $\epsilon>0$, a $2k(1+\epsilon)$-approximate matching can be maintained with an amortized update time of $O(kn^{1+1/k}\epsilon^{-1}\log \Delta)$, where $\Delta$ is the spread of $A\cup B$. Prior work on finite metric spaces gave an insertion-only data structure for $W_1$-matching. In contrast, we develop a fully dynamic approach that supports both insertions and deletions and work for all $W_p$-matchings. Together, our results extend dynamic Wasserstein matching beyond fixed-dimensional $W_1$ settings: they handle high-dimensional geometric instances, extend prior insertion-only $1$-Wasserstein matching guarantees to the fully dynamic setting, and support $p$-Wasserstein costs for every integer $p\ge 1$ in arbitrary metric spaces.

Learning effective policies for adaptive data acquisition remains challenging: posterior-based methods rely on surrogate models and posterior approximations that can be misspecified or biased, while direct policy-learning methods map from historical observations and fail to exploit available model representations, making learning harder. We introduce policy learning with belief representations (POLAR), based on the insight that optimal data acquisition depends on the observation history only through a sufficient belief state. Specifically, POLAR decouples representation learning from policy learning by leveraging pretrained predictive foundation models as belief-state encoders, training a policy head on top of their representations. This yields a simple, unified amortised policy learning framework for Bayesian experimental design, Bayesian optimisation, and active learning, differing only in the task-specific utility used to train the policy. Empirically, we find that POLAR outperforms state-of-the-art amortised methods across diverse tasks while requiring far fewer training samples, demonstrating a significant step in the scalability and efficiency of amortised data acquisition.

While animals and people tend to learn a task more quickly and reliably when they first train on simpler versions of it, curriculum learning's effectiveness in artificial settings varies, and appears substantially greater in reinforcement learning than in supervised learning. In this work, we construct and analyze a minimal model of policy learning to better understand why. We consider an open-loop, episodic target interception task whose parameters---especially how close the agent must be to the target to receive an informative reward signal---can be chosen so that the agent either receives informative rewards frequently ('dense' rewards), or rarely ('sparse' rewards). We mathematically and numerically analyze the dynamics of REINFORCE-based policy learning in both cases. In the dense reward case, curricula have only marginal benefits; in the sparse case, curricula can be required for learning to occur at all. We find that this is especially true when the agent has the conflicting goals of minimizing effort and accumulating high reward, since this can produce distinct reward landscape optima which compete with one another. Our characterization of useful curricula in this setting indicates that they address two issues: they (i) make rewards less sparse to speed up learning, and (ii) shape the reward landscape to steer agents away from bad local optima.

Knowledge conflict detection is a fundamental challenge for large language model (LLM) systems that rely on external knowledge. Existing approaches address this problem either with end-to-end supervised models or with multi-stage pipelines dependent on fixed heuristics. However, supervised models often generalize poorly to diverse conflict patterns, whereas heuristic pipelines rely on fixed rules and introduce considerable computational overhead. To overcome these limitations, we propose $\textbf{G}$rouping and $\textbf{C}$onflict $\textbf{D}$etection ($\mathsf{GCD}$), a two-stage agentic framework for knowledge conflict detection that aligns the detection process with the inherent sparsity of knowledge conflicts, thereby reducing the decision space and enabling fine-grained conflict detection. $\mathsf{GCD}$ decomposes knowledge conflict detection into two subtasks and employs two cooperative agents to address them: a Knowledge Grouping Agent (G-Agent), which creates conflict-preserving groups of textual units with potential conflicts, and an Intra-group Conflict Detection Agent (D-Agent), which detects conflicts within each group. The G-Agent's grouping policy is not fixed in advance but is jointly trained with D-Agent via a shared global reward, enabling grouping to be explicitly optimized for accurate detection rather than relying on fixed heuristics. Extensive experiments on five benchmarks demonstrate that $\mathsf{GCD}$ consistently outperforms SOTA methods, while reducing inference latency and token consumption.

Transformers predict over a representation of a sequence. The same data can be written as bytes, characters, or subword tokens, and these representations may be lossless. Yet, under a fixed context window, they need not expose the same information to the model. This raises a basic question: how does the choice of representation change what a finite-context predictor can achieve? We study this question on Markov sources and uncover two complementary phenomena. First, we observe that moving to smaller representation units can hurt prediction even when the context window is enlarged to cover the relevant source history. To explain this, we introduce *fragmentation*: a lossless recoding that replaces each source symbol by several smaller units. We prove that fragmentation can strictly increase the optimal finite-context log-loss, showing that the gap is not merely an optimization or capacity issue, but can be intrinsic to the representation. Second, we study the opposite direction: greedy tokenization, which groups source symbols into larger units. We show that tokenization can make a short token window behave like a longer source-context window, and we give a loss guarantee describing when this is achievable. The guarantee depends on how reliably token windows span the needed source history, together with the compression rate of the tokenizer. This also yields a simple diagnostic for real tokenizers: measuring how much source context a fixed token window reliably contains. Together, the two directions give a finite-context view of why representation matters for Transformer prediction.

Causal discovery is increasingly used to analyze telemetry from interactive systems, but a plausible graph does not guarantee an identifiable or reliable product decision. We study this problem for one query: whether early competitive gameplay increases Day-1 retention in a deployed Role-playing game (RPG). We propose an admissibility-first framework that treats discovered graphs as structural hypotheses, keeps only graphs that identify the target effect and have adequate treatment-control overlap, and validates admissible effects through cross-algorithm stability, placebo, subsampling, and E-value diagnostics. On real telemetry, only about one third of discovery runs support the target effect after temporal and semantic constraints. Some admissible Directed Acyclic Graphs (DAGs) converge to a positive risk-difference average treatment effect (ATE), down from a raw retention gap, and the estimate survives all refutation checks. Graph-level recovery is therefore an inadequate proxy for causal reliability; discovery pipelines for decision support should be evaluated at the query and effect level.

Knowledge editing has become a key technique for updating knowledge in large language models. However, existing methods neglect the geometric structure of the knowledge representation space, leading to two key limitations: they cannot distinguish which pre-trained knowledge should be preferentially preserved (core vs. non-core) nor which new edits are inherently difficult (hard vs. easy). To address this, we propose \textsc{GEdit}, a \textit{geometry-aware} framework built on a Low-Rank Plus Diagonal (LR+D) factor model that captures the intrinsic low-rank structure of Transformer FFN hidden representations. \textsc{GEdit} features two complementary mechanisms. First, the \textit{low-rank regularization} applies targeted protection on a low-rank core knowledge subspace. Second, the \textit{adaptive regularization} dynamically adjusts the regularization strength based on each edit request's editability: relaxing constraints for hard edits to ensure success, and tightening them for easy edits to better preserve pretrained knowledge. Extensive experiments on \texttt{GPT2-XL}, \texttt{GPT-J}, and \texttt{LLaMA-3-8B} under large-scale editing scenarios—including both mini-batch and fully sequential editing settings with up to 10,000 continuous updates—show that \textsc{GEdit} consistently and significantly outperforms state-of-the-art baselines in both editing success and knowledge preservation.

Vision–language models (VLMs) are increasingly used as automated judges for image-editing benchmarks, yet their reliability is poorly understood. Auditing these judges is difficult because real edited images lack controlled ground truth and entangle the requested change with model artifacts, unintended side effects, and ambiguous degrees of success. In this paper, we introduce EditJudge-Bench, a diagnostic benchmark of 1,500 synthetic before/after image pairs rendered in Blender, with edits parameterized across nine categories derived from real user requests and exact ground truth from known 3D scene parameters. Because every edit is generated by a known intervention on a controlled scene, EditJudge-Bench evaluates VLM judges as binary verifiers of whether an instruction was correctly executed, isolating judge perception from editor quality. Across 18 judge configurations, we find substantial reliability gaps: (a) the widely adopted VIEScore template reaches only 67.1\% AUROC with GPT-4o; (b) fine-tuned judges score 50.9–65.0\%; spatial edits such as rotation and shape remain near chance; (c) and prompt-formatting choices like image–text ordering shift performance by up to 13\%. Despite using only synthetic data, judge rankings on EditJudge-Bench correlate strongly with two real-image human-preference benchmarks ($\rho$=0.93 and $\rho$=0.77), indicating that controlled audits surface failure modes relevant to real evaluation.

Symbolic regression (SR) aims to discover compact, interpretable mathematical expressions from data. This goal inherently requires iterative refinement rather than one-shot construction. Autoregressive methods lack the ability to perform targeted corrections, so whenever structural flaws appear, the remainder of the sequence must be discarded and regenerated. Meanwhile, search-based methods rely on costly and indirect replacement through enumeration. We propose EditFlowSR, which reframes SR as iterative editing of expression trees through insertions, deletions, and substitutions. To reconcile fitting accuracy with algebraic simplicity, we introduce dual trajectory supervision. One trajectory teaches the model to construct expressions from data, while the other teaches it to algebraically simplify them, and both are unified under the shared editing framework. At inference time, EditFlowSR starts from a randomly initialized expression and progressively refines it through iterative editing, simultaneously constructing valid structure and removing redundancy. Across the standard SRBench, EditFlowSR achieves first-rank Pareto dominance in predictive accuracy, expression tree size, and time complexity. These results establish edit-based generation as a principled and effective paradigm for end-to-end symbolic regression. Code is available at \url{https://anonymous.4open.science/r/EditFlowSR_NIPS2026}.

Text-guided image editing has advanced rapidly with diffusion models and unified multimodal foundation models. However, most existing methods remain confined to single-turn settings, overlooking the more realistic scenario of multi-turn in-context editing, where users iteratively refine an image through a sequence of instructions. In this setting, a model must follow each new instruction while preserving accumulated session-level constraints, challenged by two coupled failure modes: long-context dilution, where sparse textual constraints become difficult to recover from growing interleaved image-text histories, and state contamination, where earlier editing mistakes degrade subsequent generations. We introduce Edit-R2, a novel reinforcement learning post-training framework for unified multimodal models. Edit-R2 reconstructs the operative session intent, which effectively consolidates scattered historical constraints into an explicit reasoning trace before each editing turn. It further enables multi-turn RL over both reasoning and generation through a unified objective that jointly optimizes intent reconstruction generation in discrete text space and flow-matching image generation in continuous latent space, while a trajectory filtering mechanism suppresses corrupted rollouts to stabilize training under state contamination. To support systematic evaluation, we introduce MICE-Bench, a large-scale benchmark for multi-turn in-context editing with automated metrics for instruction following (IF), content consistency (CC), and global awareness (GA) over accumulated session constraints. Experiments show that Edit-R2 substantially improves multi-turn in-context editing and achieves competitive performance compared against strong baselines.

Text-guided image editing with visual autoregressive (VAR) generators requires controlling both what the model samples and where the sampled change is written back into the image. Existing VAR editors mainly operate on token streams, features, or flat next-token logits, leaving two native structures of bitwise-residual VAR models underused: the per-bit Bernoulli prediction head and the additive multi-scale residual code field. We propose BitResEdit, a training-free editor for bitwise-residual VAR generators such as \textsc{Infinity}. BitEdit performs source-negative guidance by tilting the post-CFG per-bit log-odds along a source--target contrast computed on a shared edited prefix, then projects each update into a closed-form Bernoulli-KL trust region around the clean CFG sampler. ResEdit converts the resulting sampled bits into per-scale continuous-code residuals, gates them with a localization mask, and re-injects them through the generator's native sum-of-scales. The two components couple decision-time bit guidance with combination-time code composition, preserving masked-out latent features exactly while applying localized, scale-aware edits inside the target region. On PIE-Bench with Infinity-2B, BitResEdit achieves the strongest text alignment among same-backbone VAR editors and competitive background preservation, setting a new state of the art among training-free VAR editing methods. Ablations show that BitEdit and ResEdit play complementary roles in target alignment and background preservation. Detailed ablations show that both code-space residual editing and bitwise phrase-contrast guidance are necessary for robust autoregressive image editing.

Recent findings by \citet{cohen_gradient_2021} demonstrate that when training neural networks with full-batch gradient descent at step size $\eta$, the largest eigenvalue $\lambda_{\max}$ of the full-batch Hessian consistently stabilizes around $2/\eta$. These results have significant implications for convergence and generalization. This stabilization, however, does not occur for mini-batch stochastic gradient descent (SGD), so the implications above do not directly transfer. We show that SGD trains in a different regime we term Edge of Stochastic Stability (\textsc{EoSS}). In this regime, what stabilizes at $2/\eta$ is \emph{Batch Sharpness}: the expected directional curvature of mini-batch Hessians along their corresponding mini-batch gradients. As a consequence, $\lambda_{\max}$---which is generally smaller than \emph{Batch Sharpness}---is suppressed, aligning with the long-standing empirical observation that smaller batches and larger step sizes favor flatter minima. We further discuss implications for mathematical modeling of SGD trajectories.

Existing analyses of the edge of stability (EoS) treat it as a global property of optimization. We show that it is also selective: the stability constraint redistributes learning across subsets of the training distribution, amplifying progress on some groups while suppressing progress on others. Using a branching intervention that enters or exits the EoS regime from the same training state, we causally demonstrate this trade-off and identify two necessary conditions for a group to benefit. First, its aggregate gradient must align with the top Hessian eigenvector. We isolate this mechanism with a controlled perturbation that preserves distance but randomizes direction, destroying alignment and eliminating the advantage. Second, the group must sustain non-vanishing gradient magnitude over time. Under cross-entropy loss, gradient saturation decouples confidently classified groups, shifting the advantage to output-outliers, whose gradients persist. Together, these results show that EoS functions not only as a stability boundary, but as a mechanism governing the allocation of learning across the data distribution.

How can a population of agents self-orchestrate and self-adapt into stronger collective intelligence without centralized control? Inspired by Friedrich Hayek's economic theory of decentralized coordination in markets, we study this question through an agent economy in which agents compete via auctions for the right to act, exchange payments, and accumulate wealth from environmental rewards. These simple economic signals induce decentralized credit assignment, driving planning without global orchestration or explicit communication protocols. The population evolves through economic selection: effective agents accumulate wealth and are mutated via exploitation, while ineffective ones go bankrupt and are replaced via exploration. We show that, initialized with weak agents, the economy produces emergent multi-step reasoning strategies and outperforms stronger monolithic baselines across five agentic tasks, including mathematical reasoning, financial research, scientific research, accelerator design, and distributed-system optimization. We further provide theoretical insights into how economic dynamics shape agent behaviors, linking local incentives to long-term global performance. Our results suggest a new path to multi-agent intelligence: rather than engineering coordination, we can design decentralized incentive structures under which it automatically emerges.

In time-series forecasting, rollout errors are not merely one-step fitting discrepancies: once fed back into the evolving state, they can perturb future predictions. We formalize this effect through local rollout consistency, which compares the next prediction produced under teacher forcing with that produced under free running. In principle, this consistency can be enforced directly by tracking the residual-induced perturbation of the next state. In practice, however, the residual-to-state alignment operator is often difficult to design when the input state and output block have different dimensions, heterogeneous components, or architecture-dependent layouts. We therefore use a first-order local surrogate that penalizes the Jacobian-vector product of the predictor along the actual feedback perturbation, suppressing precisely the directions through which self-generated errors remain visible to future predictions. We further give the same loss a geometric interpretation near a predictive manifold: by controlling residual-induced directions that probe the normal bundle, the objective yields theoretical conditions under which free-running trajectories remain inside a stability tube around the manifold. Experiments on time series forecasting tasks support this view, showing that the proposed regularization improves predictive accuracy. More broadly, our results suggest a training principle for time-series forecasters and sequential predictors: a model should not only make small one-step errors, but also learn not to let its own errors echo into the future.

We introduce EchoXFlow, a clinical echocardiography dataset for learning from ultrasound in its native acquisition geometry rather than from scan-converted Cartesian videos. Existing public datasets offer limited opportunities to study cross-modal relationships between cardiac anatomy, myocardial motion, and blood flow, as Doppler is typically absent or fused as RGB overlays, and acquisitions are released after lossy vendor display processing. EchoXFlow comprises 37125 recordings from 666 routine-care examinations, preserving the timing, geometry, and modality relationships needed for physically grounded echo learning. Each recording is retained as separable modality-specific streams: temporally resolved 1D, 2D, and 3D data alongside multiple Doppler modalities, paired with a synchronized ECG. Clinical annotations span guideline-based measurements to dense 2D myocardial contours and 3D left-ventricular endocardial meshes. With its associated open-source tooling, EchoXFlow enables cross-modal, acquisition-aware learning tasks that cannot be formulated from conventional scan-converted videos alone, and serves as a testbed for 4D vision and physically grounded multi-modal learning more broadly.