arXiv subjects

Yuhao Li

Publications and source records attributed to Yuhao Li.

At least 19 recordsLinked to original sources

arXiv2026

Sybil-Proofness of Generalized Adaptive Clinching and Implications for Blockchain Mechanism Design

The adaptive clinching auction of Dobzinski, Lavi, and Nisan (FOCS'08) is a cornerstone of mechanism design with budget-constrained bidders, and it has since inspired a broad range of extensions to substantially richer market settings. Adaptive clinching enjoys compelling guarantees: it is Pareto optimal, and truthful bidding is a dominant strategy whenever budgets cannot be overreported. In many real-world environments, however, identities are cheap to create, making resilience to Sybil attacks an additional requirement. Whether the guarantees of adaptive clinching survive when a bidder can split its budget across fake identities has remained open. We introduce a generalized framework for adaptive clinching and prove that it satisfies Sybil-robust incentive compatibility: a bidder cannot benefit from splitting into any finite number of identities with arbitrary value reports, provided their aggregate reported budget does not exceed the bidder's enforceable true budget. Sybil resilience thus joins the list of ideal properties of adaptive clinching. Our work is motivated by practical problems in decentralized finance, where automated market makers (AMMs) are among the most widely used applications. We design a new AMM mechanism that extends classical adaptive clinching in two ways: it supports a two-sided market and protects against two-sided strategies, and it allows the supply function to depend on the auction state. Its one-sided special case is captured by our generalized adaptive clinching framework, and our proof techniques show that the full two-sided mechanism simultaneously satisfies Sybil-robust incentive compatibility and a relaxed notion of Pareto optimality. We argue that these guarantees are essentially the best possible via impossibility results. More broadly, our work reveals new connections between classical algorithmic game theory and modern decentralized blockchains.

cs.GT↗

arXiv2026

Tight Characterizations of AMM Mechanism Design under Weak Efficiency

Miner extractable value (MEV) poses a persistent challenge for blockchain ecosystems. It arises from the fact that a block builder can exploit its unilateral control over transaction inclusion and ordering to extract profit, potentially at users' expense. Recent work has explored consensus-aided mechanism design to mitigate MEV. This approach combines generic consensus-layer guarantees with application-layer mechanisms that use these guarantees to address application-specific MEV problems. Prior work exemplifies this approach for automated market makers (AMMs). When the consensus layer offers fair sequencing, which orders transactions according to their arrival times, their proposed mechanism achieves incentive compatibility (IC) and weak local efficiency (wLE). Informally, IC rules out profitable strategic deviations by users and builders. Meanwhile, wLE ensures that after execution, any order willing to trade at the initial price cannot benefit from further trade with the pool at the ending marginal price. In this work, we resolve the main open question left by prior works: whether IC and wLE can be achieved under plain consensus (e.g., current Ethereum) without fair sequencing. We prove that no deterministic, anonymous batch AMM can achieve both properties. The impossibility holds even when the consensus layer guarantees censorship resilience and block space is unlimited, establishing the first separation between censorship resilience and fair sequencing in the context of decentralized mechanism design. Our impossibility proof combines three strategic capabilities: creating Sybil identities, over-reporting budgets, and misreporting values. We further show the tightness of our impossibility result: removing any one from the strategy space restores feasibility under censorship resilience. Specifically, we construct a mechanism achieving IC and wLE for each restricted setting.

cs.GT↗

arXiv2026

Otter: A Provably MEV-Resilient Automated Market Maker via Surplus Redistribution

Miner extractable value (MEV) in automated market makers allows block builders to profit from transaction ordering and injected trades, imposing costs on users and contributing to builder centralization. We introduce Otter (Optimal Truthful Trading with Excess Redistribution), a two-asset batch AMM that achieves provable MEV resilience when the consensus layer provides censorship resilience and block space is uncongested. Otter makes truthful behavior a dominant strategy for both users and builders. Consequently, a builder cannot profit from strategic deviations, including reordering bids or injecting sybil bids, even when the builder is itself a trader with intrinsic value. Moreover, we show that Otter maximizes social welfare, in a strong sense, within a natural class of mechanisms satisfying the desired game-theoretic properties. To achieve these guarantees, we introduce a new paradigm called surplus redistribution, which provably prevents residual surplus from being captured as MEV by redirecting it to the broader community. Specifically, the pool's output tokens need not be distributed entirely among users in the current batch. Instead, residual surplus may be transferred to a smart contract governed by the decentralized community. The accumulated surplus can be used to benefit community members --- for example, by subsidizing transaction fees, rewarding liquidity providers, or returning assets to the pool. Our approach relies on the underlying consensus layer to provide censorship resilience. We motivate its necessity through an impossibility result showing that the desired game-theoretic guarantees become unattainable when the builder is additionally allowed to censor transactions. Thus, our results also provide a mathematically formal demonstration of how consensus-level security guarantees can fundamentally expand what is achievable at the application layer.

cs.GT↗

arXiv2026

Testing for Essential Heterogeneity in the Marginal Treatment Effects Model

We propose a test for essential heterogeneity in the marginal treatment effects model without imposing additive separability. On the observed support of the propensity score, no essential heterogeneity implies that the conditional mean of the outcome is affine in the propensity score, and this affine restriction is the sharp testable implication of the null. We implement the test using a projected Gaussian-process statistic that is Neyman-orthogonal to the first-step estimation error, so the multiplier bootstrap avoids re-estimating nuisance parameters. Monte Carlo experiments show accurate size control and good power, and, unlike the fixed-scale integrated conditional moment benchmark, the data-driven median rule retains power as the conditioning set grows in dimension. In an application to the Ugandan ELA programme, the omnibus statistic rejects the null of no essential heterogeneity for aspirations at the midline follow-up.

econ.EM↗

arXiv2026

Complexity of Autobidding Equilibria in GSP Sponsored Shopping

We revisit autobidding equilibria in GSP sponsored shopping auctions, recently proposed and studied in Dütting et al. On the existence side, we show that autobidding equilibria under both budget constraints (ABE-B) and return-on-investment constraints (ABE-R) always exist and, moreover, always admit equilibria with polynomial-size support over the allocation space. On the computational side, we show that computing an exact equilibrium is in FIXP, and that computing an approximate equilibrium is in PPAD. We also prove that computing an approximate equilibrium is PPAD-hard for both ABE-B and ABE-R. Together, these results provide a fairly comprehensive picture of the computational complexity of autobidding equilibria in GSP sponsored shopping auctions.

cs.GT↗

arXiv2026

Tarski Fixed Points in Quasi-FPT Queries

We study the query complexity of finding a Tarski fixed point over $[n]^k$. Previous work has left a large gap between $\smash{Ω(\log^2 n)}$ and $\smash{\log^{O(k)}n}$. We show that both of the previous upper and lower bounds were far from tight: for every $k\geq 3$, \[ Ω\left((\log n)^{\frac{1}{2}\lceil \log k\rceil}\right)\le \operatorname{Tarski}(n,k)\le O\left(5^k(\log n)^{\lceil \log k\rceil}\right). \] Succinctly, up to the fixed-parameter factor of $5^k$, the complexity is settled at $(\log n)^{Θ(\log k )}$. In particular, we obtain the first super-polynomial query lower bound for this problem.

cs.CC↗

arXiv2026

Data-Driven Tuning in Kernel Specification Tests: Selection Asymptotics, Validity, and Power

We develop model checks for conditional moment restrictions whose tuning parameters, the kernel bandwidth and the regularization level, are selected by cross-validation. The tests regress the estimated residuals on the covariates via kernel ridge regression (KRR), yielding a coefficient function in a reproducing kernel Hilbert space (RKHS) that vanishes under the null and is nonzero for any misspecification not annihilated by the Neyman projection. Our central contribution is a complete asymptotic theory of the data-driven tuning parameter: its limit law under the null, under fixed alternatives, and under local alternatives, and bootstrap size control under sample splitting with no convergence of the selector required. An exact energy identity shows that cross-validation maximizes the signal captured by the test (its noncentrality) and, absent signal, shrinks its noise floor. The tests are consistent against fixed alternatives, achieve $n^{-1/2}$-rate sensitivity to local alternatives, and use Neyman-orthogonal projections so the model is never re-estimated within bootstrap resamples. Simulations show size close to nominal without user-chosen tuning, power comparable to the kernel conditional moment (KCM) test at its best kernel parameter, and robustness where fixed kernel parameters fail in higher dimensions. Three empirical applications illustrate the diagnostic role of the tests; we recommend a $15$--$20\%$ selection share for the sample-splitting protocol.

econ.EM↗

arXiv2026

MinkowskiPE: Minkowski Positional Encoding for Spatiotemporal Perception

Modeling spatiotemporal coupling is a key challenge in building physical intelligence across scales, from microscopic to macroscopic. Existing models capture such structure broadly through physics-motivated dynamical formulations or learning-motivated architectures. The former provide stronger priors but may constrain flexibility, whereas the latter are more flexible but leave the spatiotemporal coupling largely implicit. We therefore seek an approach that combines flexible learning with an explicit geometric bias for jointly modeling time and space. To this end, we propose Minkowski Positional Encoding (MinkowskiPE), which uses joint temporal and spatial coordinates to parameterize Lorentz transformations applied to query and key features. With MinkowskiPE, the query-key attention score depends on position only through the relative spacetime displacement between the two tokens and is therefore invariant to global translation of the coordinates. This paradigm retains the standard dot-product attention interface and remains compatible with efficient attention implementations. We evaluate MinkowskiPE on microscopic molecular dynamics and macroscopic video prediction tasks, achieving the best results on all nine multi-trajectory molecular evaluations and reducing KTH video-prediction MSE by 9.9% relative to the best baseline while using roughly one-tenth as many parameters.

cs.LG↗

arXiv2026

The Linear Representation Hypothesis Needs a Group Action

To make claims about representations that generalize beyond a particular trained model, we need to specify when two representations should count as equivalent. The Linear Representation Hypothesis is often discussed without making this equivalence explicit. Different notions of equivalence preserve different structures, so metrics, probes, and interventions that appear to study the same representation may in fact correspond to different hypotheses. We therefore argue that the Linear Representation Hypothesis is not one hypothesis but a family of claims distinguished by representation equivalence. We formalize this idea using group actions, specifying the representation object, the procedure that produces it, and the property ultimately asserted, while accounting for equivalences imposed by the model architecture. This framework clarifies how assumptions can change across metrics, reading points, and analysis stages, and we use it to audit common representation quantities and recent interpretability analyses.

cs.LG↗

arXiv2026

Moment Propagation

Mean-field variational Bayes is a fast and scalable approach to approximate Bayesian inference, but its independence assumptions often lead to underestimated posterior uncertainty. We introduce moment propagation (MP), a framework for improving marginal posterior approximations by propagating conditional posterior moment information between parameter blocks and matching these moments within convenient approximating families. In conjugate settings, MP identifies variance terms omitted by mean-field approximations and uses them to construct corrected marginal updates. We develop both mean-field and Gaussian variants of MP, the latter using conditional Gaussian structure to obtain low-dimensional local updates for non-conjugate models. We establish consistency and asymptotic variance correctness for two-component models, and show that MP recovers exact marginal posteriors for linear regression and multivariate normal models. We also derive algorithms for multivariate normal models with missing data, probit regression, generalized linear models, and Bayesian Lasso regression. Numerical experiments show that MP, particularly Gaussian MP, can substantially improve marginal posterior accuracy over standard variational approximations while remaining much faster than MCMC.

stat.CO↗

arXiv2026

Kernel Minimum Distance Estimation and Testing with Conditional Moment Restrictions: A Unified Framework

We propose a unified Kernel Minimum Distance (KMD) framework for estimating and testing models defined by conditional moment restrictions. By embedding conditional moments into a Reproducing Kernel Hilbert Space (RKHS), we construct a closed-form $V$-statistic objective function that quantifies the distance from the restrictions. We establish the $\sqrt{n}$-consistency and asymptotic normality of the associated minimum distance estimator. Within this framework, the minimized objective function naturally yields a consistent omnibus specification test. Unlike projection-based methods that require auxiliary nonparametric estimation for Neyman orthogonalization, our test inherently captures the estimation effect via a projected kernel structure. We derive asymptotic properties of the test statistics under the null hypothesis, the alternative hypothesis, and a sequence of local alternatives converging to the null at the parametric rate $n^{-1/2}$. The validity of a computationally simple multiplier bootstrap is established to facilitate inference. Simulation results demonstrate robust finite-sample performance, and the framework is illustrated by analyzing Engel curves using UK Family Expenditure Survey data.

econ.EM↗

arXiv2026

Kernel Two-Sample Testing via Directional Components Analysis

Standard kernel two-sample tests, such as those based on the Maximum Mean Discrepancy (MMD), aggregate squared differences across all directions in a Reproducing Kernel Hilbert Space (RKHS). However, in finite samples, trailing directional components are noisy, which degrades test power. We propose a novel kernel-based test that resolves this by truncating the spectral decomposition of the MMD, retaining only the well-estimated leading eigen-directions. By aggregating these robust components, our method achieves superior power and robustness, particularly in high-dimensional and unbalanced settings. Furthermore, we introduce a computationally efficient parametric bootstrap procedure for approximating critical values, which is theoretically justified and significantly faster than permutation-based alternatives. Extensive simulations and empirical studies demonstrate that our method maintains strict Type I error control while delivering higher power than existing MMD-based tests.

stat.ME↗

arXiv2026

Curvature-Guided Geometric Representation for Protein-Ligand Binding Affinity Prediction

Protein-ligand binding affinity (PLA) prediction is critical in drug discovery. Despite the notable advancements in machine learning-based approaches, existing methods struggle to jointly characterize local geometric organization and globally coordinated cross-molecular interactions, limiting their ability to model complex binding mechanisms. Here, we propose RicciBind, a geometric representation framework that integrates curvature-guided hierarchical structure learning with optimal transport (OT)-based cross-domain alignment to model molecular interactions. Specifically, RicciBind leverages Ricci curvature to capture local interaction tightness within molecular structures, enhancing structural awareness and organizing atomic interactions into curvature-aware hierarchical representations. An OT-based cluster matching mechanism then aligns protein and ligand clusters across heterogeneous domains under geometric constraints, enabling globally consistent correspondences and revealing higher-order interaction patterns beyond local neighborhoods. By coupling curvature-guided structure encoding with OT-driven cross-domain alignment, RicciBind effectively models complex interaction semantics and substantially improves both the accuracy and interpretability of binding affinity prediction. Extensive experiments demonstrate that RicciBind achieved superior predictive performance and generalization across PLA benchmarks and virtual screening tasks. Ablation studies further confirmed the essential role of Ricci curvature in enhancing molecular interaction representations.

cs.LG↗

arXiv2026

A Trilemma in AMM Mechanism Design

Blockchains have popularized the Automated Market Makers (AMMs), where users trade crypto-assets directly with a smart contract, governed by a pricing function embedded in the contract's code. Today, users of AMMs are often forced to accept unfavorable prices due to widespread front-running and back-running attacks, commonly known as Miner Extractable Value (MEV). Several earlier works show impossibility results suggesting that completely removing MEV at the consensus layer is impossible, partly because the consensus layer is agnostic of application-level semantics. For this reason, more recent works have advocated mechanism design approaches at the application (i.e., smart contract) level. We study a natural two-asset AMM mechanism design problem recently initiated and explored in prior work by Chan, Wu, and Shi, in which they proposed a mechanism that satisfies a surprisingly strong notion of incentive compatibility (IC), under the consensus assumption that the underlying blockchain provides sequencing fairness. In this paper, we investigate the (in)feasibility of simultaneously achieving IC and other desirable properties such as weak local efficiency (wLE) and uniform pricing (UP). At a high level, wLE requires that the mechanism should not leave any unfulfilled demand from users whose asking prices are not overly restrictive, and whose orders could have been executed directly against the pool. UP requires that all orders that get (partially) executed must trade at the same exchange rate. We unveil the underlying mathematical structure of AMM mechanism design, and our main results can be summarized as a trilemma-style theorem: among the desirable properties IC, wLE, and UP, any two out of three are possible, but no mechanism can satisfy all three.

cs.GT↗

arXiv2026

Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks

Deep reasoning is fundamental for solving complex tasks, especially in vision-centric scenarios that demand sequential, multimodal understanding. However, existing benchmarks typically evaluate agents with fully synthetic, single-turn queries, limited visual modalities, and lack a framework to assess reasoning quality over multiple steps as required in real-world settings. To address this, we introduce Agent-X, a large-scale benchmark for evaluating vision-centric agents multi-step and deep reasoning capabilities in real-world, multimodal settings. Agent- X features 828 agentic tasks with authentic visual contexts, including images, multi-image comparisons, videos, and instructional text. These tasks span six major agentic environments: general visual reasoning, web browsing, security and surveillance, autonomous driving, sports, and math reasoning. Our benchmark requires agents to integrate tool use with explicit, stepwise decision-making in these diverse settings. In addition, we propose a fine-grained, step-level evaluation framework that assesses the correctness and logical coherence of each reasoning step and the effectiveness of tool usage throughout the task. Our results reveal that even the best-performing models, including GPT, Gemini, and Qwen families, struggle to solve multi-step vision tasks, achieving less than 50% full-chain success. These findings highlight key bottlenecks in current LMM reasoning and tool-use capabilities and identify future research directions in vision-centric agentic reasoning models. Our data and code are publicly available at https://github.com/mbzuai-oryx/Agent-X

cs.CV↗

arXiv2026

On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective

Debates about large language model post-training often treat supervised fine-tuning (SFT) as imitation and reinforcement learning (RL) as discovery. But this distinction is too coarse. What matters is whether a training procedure increases the probability of behaviors the pretrained model could already produce, or whether it changes what the model can practically reach. We argue that post-training research should distinguish between capability elicitation and capability creation. We make this distinction operational by introducing the notion of accessible support: the set of behaviors that a model can practically produce under finite budgets. Post-training that reweights behaviors within this support is capability elicitation; whereas changing the support itself corresponds to capability creation. We develop this argument through a free-energy view of post-training. SFT and RL can both be seen as reweighting a pretrained reference distribution, only with different external signals. Demonstration signals define low-energy behavior for SFT, and reward signals define low-energy behavior for RL. When the update remains close to the base model, the main effect is local reweighting, not capability creation. Within this framework, the central question is no longer whether post-training is framed as SFT or RL, but whether it reweights behaviors already within reach, or instead expands the model's reachable behavioral space through search, interaction, tool use, or the incorporation of new information.

cs.AI↗

arXiv2026

HER: Human-like Reasoning and Reinforcement Learning for LLM Role-playing

LLM role-playing, i.e., using LLMs to simulate specific personas, has emerged as a key capability in various applications, such as companionship, content creation and digital games. While current models effectively capture character tones and knowledge, simulating the inner thoughts behind their behaviors remains a challenge. Towards cognitive simulation in LLM role-play, previous efforts mainly suffer from two deficiencies: lacking data with high-quality reasoning traces, and lacking reliable reward signals aligned with human preferences. In this paper, we propose HER, a unified framework for cognitive-level persona simulation. HER introduces dual-layer thinking, which distinguishes characters' first-person thinking from LLMs' third-person thinking. To bridge these gaps, we curate reasoning-augmented role-playing data via reverse engineering, and construct human-aligned principles and reward models. Leveraging these resources, we train HER models based on Qwen3-32B via supervised and reinforcement learning. Extensive experiments validate the effectiveness of our approach. Notably, our models significantly outperform the Qwen3-32B baseline, achieving a 30.26 improvement on the CoSER benchmark and a 14.97% gain on the Minimax Role-Play Bench. Our datasets, principles, and models are released to facilitate future research.

cs.LG↗

arXiv2026

Abstract Sim2Real through Approximate Information States

In recent years, reinforcement learning (RL) has shown remarkable success in robotics when a fast and accurate simulator is available for a given task. When using RL and simulation, more simulator realism is generally beneficial but becomes harder to obtain as robots are deployed in increasingly complex and widescale domains. In such settings, simulators will likely fail to model all relevant details of a given target task and this observation motivates the study of sim2real with simulators that leave out key task details. In this paper, we formalize and study the abstract sim2real problem: given an abstract simulator that models a target task at a coarse level of abstraction, how can we train a policy with RL in the abstract simulator and successfully transfer it to the real-world? Our first contribution is to formalize this problem using the language of state abstraction from the RL literature. This framing shows that an abstract simulator can be grounded to match the target task if the grounded abstract dynamics take the history of states into account. Based on the formalism, we then introduce a method that uses real-world task data to correct the dynamics of the abstract simulator. We then show that this method enables successful policy transfer both in sim2sim and sim2real evaluation.

cs.RO↗