Search arXiv⌕ Search

SEARCH · Search arXiv

Search Search arXiv

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,279 records · Page 71Linked to original sources
arXiv2026

ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents

Large Language Model (LLM) agents increasingly rely on external memory to support long-horizon reasoning and decision making. Existing memory systems typically retrieve historical trajectories or summaries as independent context fragments, overlooking the procedural dependencies underlying multi-step execution. As memory scales, such flat retrieval introduces context fragmentation and cross-task interference, leading to structurally inconsistent reasoning trajectories. We propose ActiveMem, a hierarchical memory framework that recursively organizes agent experiences into dependency-aware latent execution trees. ActiveMem abstracts trajectories into reusable subtask nodes while explicitly preserving execution transitions, enabling coherent reasoning-path retrieval conditioned on the current execution state. To support continual adaptation, ActiveMem further learns dynamic memory expansion, retrieval, and pruning policies through reinforcement learning. Experiments across various agent benchmarks demonstrate that ActiveMem consistently improves task completion, reasoning stability, and memory efficiency over existing memory-based agents. Moreover, ActiveMem enables compact open-weight models to achieve competitive performance with substantially larger proprietary systems.

cs.AI↗
arXiv2026

LSTMem: Hierarchical Long Short-Term Online Memory for Large Language Models

Large language models increasingly serve as long-horizon assistants and agents, where they must both accumulate information across interactions and make the relevant parts available when later requests depend on them. Existing compact online memories typically use a single persistent state both to accumulate history and to serve readout, so what the memory stores cannot be controlled separately from what it exposes to the current computation. We propose LSTMem, an LSTM-inspired online memory that instead equips each layer of a frozen LLM with two matrix-valued states: a cell state that accumulates history and a hidden state whose readouts correct the backbone's attention. Input and forget gates control what the cell stores, while an output gate separately controls what the cell exposes through the hidden state. LSTMem further connects memory across depth through forward hidden-state propagation and block-end feedback, and uses higher-layer reconstruction gradients to refine lower-layer cell states before rebuilding hidden states from shallow to deep layers. Across memory benchmarks on Qwen3-4B-Instruct, LSTMem consistently improves MemoryAgentBench, LoCoMo, and HotpotQA over the plain backbone. Comparisons further show that the LSTM-based memory formulation outperforms an associative-memory counterpart, while removing cross-layer hidden-memory propagation degrades performance. These results demonstrate the benefits of separating memory accumulation from memory expression and organizing memory hierarchically across model depth. The code is available at https://github.com/Longchentong/LSTMem.

cs.AI↗
arXiv2026

A Hopf Algebraic Theory of the Quantum Magnusian

We develop a Hopf-algebraic theory of graph coefficients arising in the quantum Magnus expansion. At the classical level, the expansion is governed by directed trees, whereas its quantum counterpart involves loop graphs, multiple edges, and several types of edge data. We introduce a contraction Hopf algebra of mixed quivers containing these diagrams and closed under contraction. On this algebra we define a character $e$ from normalized linear-extension data, directed-tadpole factors, and a prescription for undirected edges, and let $ω$ be its convolution inverse. Our main result is a universal closed formula for the connected function $ω_c$ on every finite mixed quiver, expressed as a finite sum over ordered set partitions of the vertices. We also construct a weighted vertex-splitting Hopf structure on the unlocalized mixed-quiver algebra, with convolution $\star_V$ and canonical edge-weight character $ζ_V$, such that $ω_c=\log_{\star_V}ζ_V$. Compatibility of the two coproducts and an explicit identity between their characters explain this equality. We further prove that the two-vertex contraction identity, tadpole factorization, vanishing on disconnected graphs, and boundary data uniquely determine $ω_c$. For acyclic quivers, the formula reduces to a permutation formula with coefficients depending only on descent numbers. We derive this formula independently from the quantum Magnus expansion using operator products and Wick contractions, showing that the physical graph coefficients are governed by the convolution structure of the contraction Hopf algebra. Further consequences include a refined quantum Murua formula and orientation-sum identities related to Tutte and chromatic polynomials.

math.CO↗
arXiv2026

Marginal Calibration Masks Generator Differences in Held-Out Evaluation of Multivariate Time Series, with an Application to ERA5 Weather Sequences

Generative models for multivariate time series are routinely evaluated with pooled distributional metrics computed after marginal calibration. We show that, under a held-out protocol, the calibrated Kolmogorov-Smirnov (KS) distance differs from a real-versus-real floor by at most the distance of the calibrated sample to the training block, so that it carries little information about the generator. We audit a three-discriminator GAN with evolutionary weight adaptation (StatD2GAN) against re-implemented RCGAN and TimeGAN baselines on ERA5 weather sequences under a pre-specified protocol: the final two calendar years are held out behind a 168 hour embargo, calibration is fitted on the training block only, and the decision rules were frozen in advance. Evidence comes from 25 matched (location, seed) pairs across five locations, tested with Holm-corrected Wilcoxon signed-rank tests. First, isotonic calibration drives KS to a median of 1.019 times a per-location noise-and-shift floor, and neither baseline is distinguishable from StatD2GAN on it. Second, the sorted-representation discriminator is the only component whose removal significantly degrades cross-variable dependence (Kendall tau MAE +0.116, Holm p = 0.002); a rank-transformed variant also degrades it, consistent with quantile supervision rather than copula matching. Third, calibration raises physical constraint violations; projection removes them, but the largest marginal change (0.0110, Lhasa) exceeds the pre-specified tolerance in three locations. Fourth, the re-implemented TimeGAN is significantly better than StatD2GAN on Kendall tau and ACF MAE, RCGAN is worse on Kendall tau MAE and constraint violations, and the pre-specified collapse criterion was not met. We recommend floor-referenced marginal evaluation, matched-pair testing, and dependence and sequence-level metrics.

cs.LG↗
arXiv2026

Stability of 3D Stratified Plane Couette Flow: A New Lift-up Mechanism and Potential Vorticity

Since the pioneering work of Taylor and Goldstein in 1931 laid the foundation for the study of parallel stratified shear flows, non-parallel configurations have received comparatively little attention. We study the stability of the 3D Boussinesq system near the plane Couette flow $V^*=(y,0,0)$ under the vertically stratified background $η^*=αz$. The shear and stratification directions are perpendicular. This geometry produces two structural effects that are absent in the parallel configuration: an additional nonlocal coupling in the non-zero-mode system, which obstructs a directly useful symmetrization, and a wave-shear coupled lift-up mechanism for the zero modes. The $L^p$ transient growth due to the lift-up effect can be suppressed either by sufficiently strong stratification (for $p>2$) or under suitable assumptions on the initial data (for $p\geq 2$). To study the nonzero-mode system, we introduce a reformulation based on the linearized potential vorticity (PV), a fundamental quantity in geophysical fluid dynamics, to recover a better energy structure. We establish the inviscid damping of $u_{1,\neq}$ and $u_{2,\neq}$. We further apply the linearized PV formulation to study the linear stability of the tilted Couette flow. Finally, we use this formulation to prove the nonlinear stability for perturbations on $\mathbb{T}\times\mathbb{R}\times\mathbb{T}$.

math.AP↗
arXiv2026

StructSim: Measuring Idea Similarity at Scale Through Structural Representation

Measuring idea similarity is fundamental to creativity evaluation, especially as LLMs enable idea generation at increasing scale. However, text embeddings collapse an idea into a single vector, making it difficult to capture structural similarity, including partial overlap across core and supporting components and differences across levels of abstraction. We introduce a shared structural representation that decomposes ideas into purpose, mechanism, and implementation components and organizes related components in a multi-layer concept graph. From this representation, we define measures of pairwise similarity and set-level mechanism coverage for assessing idea diversity. We evaluate our approach using controlled idea triples and assessments from 12 experts, focusing on differences in core mechanisms, implementations, and supporting components. Our method improves alignment with expert judgments of structural similarity by 31% over the embedding baseline and better reflects expert assessments of idea set coverage, supporting scalable evaluation of idea similarity and diversity.

cs.HC↗
arXiv2026

CoSec: Benchmarking Agent Security in Communities

LLM agents operate in persistent collaborative environments involving multiple users, communities, memories, files, and tools. Community boundaries may remain fixed or evolve with changes in membership, roles, composition, and relationships. Agents must complete legitimate tasks and prevent unauthorized disclosure of protected information. Existing evaluations do not fully examine these risks in agent systems. We introduce \textbf{CoSec}, an executable benchmark for evaluating privacy and authorization enforcement in LLM agent systems operating within and across communities. CoSec contains 208 canonical scenarios spanning fixed and evolving boundaries, protected information belonging to the agent owner or other participants, and attacks through dialogue, environmental content, persistent memory, and composed workflows. CoSec executes complete agent systems with persistent sessions, memory, files and tools. It verifies information flows against the active authorization state using execution traces and artifacts. Across harness and model configurations, agents frequently complete benign tasks but violate privacy and authorization boundaries. Privacy behavior varies across harnesses, attack surfaces, and community states, revealing how memory, files, tools, and workflows can carry protected information beyond its authorized scope. These findings show that task utility does not imply privacy or authorization compliance and that authorization in community settings remains an unresolved security challenge for persistent LLM agents.

cs.CR↗
arXiv2026

DoAtlas-2: A Foundation for Self-Evolving Causal Biomedical Discovery

We introduce DoAtlas-2, a foundation for self-evolving causal biomedical discovery that organizes knowledge around causal mechanisms and advances through external evidence from human populations. DoAtlas-2 integrates 771 research resources covering more than 720,000 participants in 48 countries, from longitudinal clinical phenotypes, medical imaging, and continuous physiological signals to eight molecular layers, together with an evidence network of approximately 4.7 million literature-derived records over 93,566 concepts and 149,383 candidate causal relations. DoAtlas-2 autonomously formulates research questions from evidence gaps and unresolved mechanisms, prespecifies their causal designs, and generates validated analyses. Supporting, challenging, and unresolved results continuously revise mechanistic interpretations, the causal evidence state, and the discovery frontier, so that DoAtlas-2 self-evolves within a closed loop of hypothesis generation, empirical testing, and renewed discovery. DoAtlas-2 has systematically evaluated 2,031 research questions. In the Human Phenotype Project (HPP), it formulated 4,014 candidate pathway questions across vascular, early-glycemic, and hepatic-metabolic systems, and screening of the first 1,079 yielded statistical support for 756. Representative studies identify blood pressure as a convergence node linking adiposity, hepatic, and lipid phenotypes to vascular outcomes, and show that an adiposity-inflammation-blood-pressure pathway is largely attenuated by joint adjustment for body mass index (BMI) and smoking. The discovered vascular network constitutes a completely interpretable predictive foundation, admitting exact attribution of every prediction and closed-form mediation effects. DoAtlas-2 thereby unifies causal mechanism discovery, population-evidence testing, and interpretable prediction within one continuously evolving foundation.

cs.AI↗
arXiv2026

Right Facts, Wrong Rule: Scope-Transfer Poisoning of Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) systems use external documents for facts and rules that guide answers. We study rule-scope misapplication, where a rule valid within its documented scope is applied to a question outside that scope. Correct facts can then support an incorrect answer even when the rule's boundary is visible. This exposes the limits of document-validity checks. We introduce LENS (Local Evidence, Non-local Steering), an LLM-agent-based attack framework for constructing auditable instances of this failure. Its Planner identifies a grounded rule, its legitimate use, intended misuse, and document roles. Proposer and Verifier agents construct and repair attack documents using local retrieval and reader feedback while preserving factual support and explicit scope boundaries. Across four QA benchmarks and three target generators, LENS achieves 47.1% target-error rate over 4,000 fixed attempts, including construction failures. A separate fixed-attempt audit yields 36.1% joint target-error rate under independent validity checks. Fact-preserving interventions link errors to rule scope. For 76.2% of target-matching default responses, the same model correctly judges the rule inapplicable when explicitly probed. With matched information and budgets, an applicability audit reduces target error from 70.8% to 16.4%, compared with 31.2% for task-alignment checking. These findings motivate checking and enforcing rule applicability during answer generation.

cs.CR↗
arXiv2026

Hyper Algorithm Design Agent: Evolving Learnable Optimizer from Zero

Meta-Black-Box Optimization (MetaBBO) is one of the highlights in the recent AI for Optimization trend. This paradigm's bi-level workflow leverages the learnable algorithm design policy at meta level to ensure the performance and generalization improvement on the low-level optimization task. While MetaBBO helps advance the performance lower bound of the resulted optimization system, it is currently handcrafted and customized case by case to adapt different optimization problems, which inevitably introduces inherent subjectivity and hence restricts the performance upper bound and usability in practice. In this paper, we address this issue by regarding MetaBBO's design loop as coding task, where we could introduce openendedness into MetaBBO with recursive self-improvement capability of advanced coding agents. Specifically, we propose a dual-agent framework: i) a task agent continuously refines the codebase of a target MetaBBO approach through code evolution; ii) a hyper agent progressively modifies the task agent and itself to provide open-ended design behavior; iii) the evolved MetaBBO codebase is evaluated and all in-execution information is fed back to the agents for recursive self-referential improvement. As a result, given a naive MetaBBO template, our framework automates a design evolution and finds novel variants superior to up-to-date human-made MetaBBO baselines. Surprisingly, the experimental results also demonstrate that our framework supports fast adaption across different optimization domains. Solid interpretation analysis further reveals interesting design principles emerge in such open-ended process. This work serves as the first exploration on automating design of complex learning-assisted optimization algorithms.

cs.AI↗
arXiv2026

Just Initialize: A Training-Free Initialization Component for Large-Scale Routing Optimization

Large-scale routing problems are difficult to solve efficiently as their search spaces grow rapidly with problem size. Existing approaches primarily improve the optimization procedure itself, often at increasing computational cost. We instead shift the focus to a useful initialization that can be refined into a high-quality solution with limited downstream refinement. We propose Just Initialize, a training-free and solver-agnostic initialization component for large-scale routing optimization. Just Initialize compresses a large routing instance into a compact surrogate space, optimizes its global routing structure, and recovers the resulting solution as an optimization-friendly starting point in the original space. Extensive experiments on Traveling Salesman Problems (TSPs), Capacitated Vehicle Routing Problems (CVRPs), Vehicle Routing Problems with Time Windows (VRPTWs), and Prize-Collecting Traveling Salesman Problems (PCTSPs) demonstrate that Just Initialize achieves high-quality solutions comparable to or better than state-of-the-art methods while substantially reducing computational cost across instances ranging from 1K to 100K nodes, including an average speedup of approximately 70$\times$, sub-second runtimes on 10K-node instances, and runtimes within tens of seconds on 100K-node instances.

cs.AI↗
arXiv2026

Binarization Flattens the Score Space

Large language model (LLM) judges are often used as rewards to train policies on objectives that deterministic verifiers cannot capture. However, these rewards are often collapsed to pass/fail ({0, 1}), which reports the verdict but not how well a response met each criterion. We model each pass/fail verdict as a score on an unreported scale, compared with one cutoff. A stretch of that scale moves every score proportionally toward or away from the cutoff, but never across it, so no verdict changes. A policy is therefore free to apply any stretch without changing anything the panel reports. Under a joint-Gaussian model, a third grade adds a second threshold and removes this affine stretch ambiguity. On MATH and SciBench outputs from one seven-criterion judge, all 14 constructed criterionwise stretches were invisible after binarization but visible with three grades. At $n=1{,}024$, a test given both population laws had at least 96.5% power at a $1.5\times$ stress. Retaining grades closes one blind spot created by binarization, but verdicts alone remain insufficient as some changes are still indistinguishable from genuine improvement. These include arbitrary within-grade changes and fixed-covariance, loading-aligned mean shifts -- the signature of a sycophancy-shaped lift the panel reads as competence. The shared-factor reference approximation fit MATH and SciBench but not HealthBench, delineating its empirical scope. We recommend keeping at least three grades (for example, asking the judge whether each criterion is fully, partially, or not met and rewarding {0, 0.5, 1}), and externally validating gains along the remaining direction, which no finer scale removes.

cs.LG↗
arXiv2026

Matrix-Vector Complexity of Low-Rank Approximation

We establish matching polynomial query bounds for low-rank approximation from exact matrix--vector products. Given an unknown matrix $A\in\mathbb{R}^{m\times n}$, at each step a randomized algorithm chooses either $v\in\mathbb{R}^n$ and receives $Av$, or $u\in\mathbb{R}^m$ and receives $A^\top u$. The choice may depend measurably on all previous queries and replies and on the algorithm's private randomness; each vector product costs one query. The output is a rank-$k$ right projector with Schatten-$p$ residual at most $1+\varepsilon$ times optimal. Write $N=\min\{m,n\}$ and let $Q_p^*$ denote the worst-case query budget for success probability $2/3$ on every input. For every $1\le k<N$ and sufficiently small $\varepsilon$, our lower bounds, combined with existing Krylov upper bounds, give $Q_p^*=\widetildeΘ\!\left(\min\{N,k\min\{p^{1/6}\varepsilon^{-1/3},\varepsilon^{-1/2}\}\}\right)$ $(2\le p<\infty)$, $Q_\infty^*=\widetildeΘ\!\left(\min\{N,k\varepsilon^{-1/2}\}\right)$. These bounds have universal constants and allow $p$ to vary with the problem parameters, identifying the transition at $p\varepsilon\asymp1$. A complementary result for each fixed $1\le p<2$ gives $\widetildeΘ_p(\min\{N,k\varepsilon^{-1/3}\})$, with constants and an accuracy threshold that may depend on $p$. Together, the results recover this fixed-norm rate for every fixed finite $p$, supplying the multiplicative rank dependence missing from previous lower bounds. Tildes suppress logarithmic factors. The proof extends adaptive Wishart deferred decisions to a rectangular factor with a $k$-dimensional nullspace. Posterior overlap gives a short fixed-norm argument, while persistence of small compression eigenvalues controls growing $p$ and the spectral endpoint. Exact range recovery handles target costs of order $k$; the Wishart family covers the remaining regimes.

math.OC↗
arXiv2026

Positive Cubature Compression and Optimal Hyperinterpolation Stability on a Conic Surface

We study the computational cost and stability of positive cubature and hyperinterpolation on a truncated conic surface with a Jacobi-type weight. A classical quadratic disk-to-cone map is formulated as a measure-preserving $\mathbb Z_2$ quotient. It identifies degree-$m$ conic traces with even disk polynomials of degree at most $2m$ and gives a two-way correspondence between positive degree-$2n$ cone cubature and centrally symmetric positive degree-$(4n+1)$ disk cubature. Consequently, Möller's lower bound and the node excess above it transfer exactly, producing compressed non-product cone rules from near-minimal disk formulas. The quotient also transfers reproducing kernels and hyperinterpolation operators. Every positive degree-$2n$ cone rule yields an exact $L^2$ sampling isometry on $Π_n(V)$, so the weighted sampling matrix has condition number one, independently of the number and geometry of its nodes. The associated Lebesgue constants satisfy the rule-independent bound $Λ_{n,γ}^V\ge c_γn^{γ+1}$ for every $γ>-1$. For nonnegative half-integer $γ$, the matching upper bound holds, giving the sharp universal order $Λ_{n,γ}^V\asymp n^{γ+1}$; in particular, $Λ_n^V\asymp n$ when $γ=0$. Thus node compression preserves exact $L^2$ conditioning and, in this parameter range, optimal-order $L^\infty$ stability. Low-degree near-minimal rules and higher-degree optimized disk schemes illustrate the reduction in sampling cost.

math.CA↗
arXiv2026

Channel-Dependent State Space Model for Multivariate Time Series Forecasting

Multivariate time series forecasting (MTSF) is critical across many real-world domains. Existing deep learning approaches fall into two paradigms with distinct limitations: channel-independent (CI) methods unconditionally ignore cross-variable dependencies and model only temporal dynamics, while channel-dependent (CD) methods consider both but typically rely on architectural compromises to mitigate overfitting and computational overhead. We therefore propose Chameleon, a specialized CD state space model (SSM) that enables data-dependent, fine-grained interactions across variables while scaling linearly with their number. By connecting selective SSMs with the Kalman filter, we leverage the missing measurement update in the former for cross-variable modeling while preserving the SSM backbone for robust temporal modeling. We further identify favorable inductive biases of GatedDeltaNet for time series, adapt it as our backbone, and improve generalization through additional techniques, including a previously unexplored stochastic perturbation of reversible instance normalization. On strongly dependent ODE and PEMS datasets, Chameleon achieves the best MSE and MAE across all settings, while its CI ablation and prior CD methods incur 61-178% higher MSE on average. Across 28 standard benchmark settings, Chameleon also achieves better MSE and MAE than each baseline in at least 27 and 22 cases, respectively. Training-time and peak-memory analyses on Traffic and ETT further demonstrate competitive efficiency and favorable memory scalability across different variable counts.

cs.LG↗
arXiv2026

FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation

A critical prerequisite of generative recommendation is designing semantic identifiers (SIDs) that are both scalable to large item sets and efficiently learnable. Existing SID learning methods fundamentally rely on Top-1 hard assignment during vector quantization. While heuristic strategies -- such as clustering-based initialization or forced post-hoc collision resolution -- can artificially inflate codebook coverage, they often disrupt end-to-end semantic alignment and fail to address the underlying optimization bottleneck: sparse gradient propagation. In standard Top-1 assignment, gradients concentrate on a narrow subset of frequently selected codewords, leaving the majority inherently under-trained and causing severe SID collisions. To overcome this limitation natively without relying on complex initialization priors, we propose FineSID, a unified quantization framework that moves beyond Top-1 assignment by enabling fine-grained gradient propagation across the entire codebook. Instead of updating only a single selected codeword, FineSID distributes learning signals to all codewords in a soft, differentiable manner. This design promotes globally balanced codebook optimization while strictly preserving semantic consistency, effectively alleviating SID collisions and stabilizing training in large, high-dimensional codebooks. Extensive experiments on multiple public benchmarks demonstrate that FineSID is robust to initialization configurations and consistently improves both codebook utilization and recommendation accuracy. Our work provides a principled, initialization-agnostic solution for semantic identifier learning, advancing the practicality of generative recommendation.

cs.AI↗
arXiv2026

FairDiff: Mitigating the Self-Reinforcing Matthew Effect in Diffusion Recommender Models

While the "Matthew Effect" and filter bubbles are widely recognized outcome-level biases in recommender systems, we reveal that Diffusion Recommender Models (DRMs) uniquely compound this issue through their generative dynamics. Rather than merely inheriting data imbalances, DRMs trigger a self-reinforcing amplification of popularity bias. We identify that this phenomenon is driven by two compounding mechanisms. First, while optimization loss is universally dominated by high-frequency items across recommenders, DRMs suffer from a unique structural prior mismatch during generation. Because the forward terminal distribution of long-tailed data deviates significantly from the standard Gaussian prior, reverse sampling trajectories inherently collapse toward high-density popular items, fundamentally suppressing niche item generation. To dismantle this self-reinforcing loop, we propose FairDiff, a plug-and-play fairness-aware diffusion framework. To overcome the popularity-dominated loss, we introduce Popularity Condition Guidance (PCG). Rather than altering the training objective, PCG acts as an inference-time distributional reweighting mechanism, mathematically reshaping the score-based gradient field to penalize high-popularity regions and guide trajectories toward niche semantics. Furthermore, we design a Semantic Calibration (SC) Module to bridge the prior mismatch, aligning the forward and reverse distributions via one-step optimal transport. Comprehensive evaluations demonstrate that FairDiff achieves state-of-the-art performance while effectively mitigating the self-reinforcing Matthew Effect, highlighting its value as a general framework for DRMs.

cs.AI↗
arXiv2026

WeLike2Party! In-Context Motion Transfer for Multi-Human Image Animation

Human image animation aims to transfer motion from a driving video to subjects in a reference image. Despite remarkable progress in video generation, achieving high-fidelity animation of multiple interacting subjects remains a challenge. Many existing approaches rely on explicit motion representations such as 2D skeletons or parametric body meshes and struggle to preserve identity-motion binding under inter-person occlusion. To address this limitation, we propose WeLike2Party, a multi-human animation framework built on direct in-context video conditioning without explicit pose or mesh extraction at inference. We further introduce Reference Asymmetric RoPE Conditioning to preserve fine-grained appearance details, and Identity Binding Supervision to associate each reference identity with its intended motion trajectory. To support cross-identity training, we construct MotionTwin, a large-scale synthetic dataset comprising 14.4K cross-identity video pairs with shared subject and camera motions, totaling 84.3 hours of photorealistic video. We additionally present MotionTwin-Bench, a cross-identity benchmark specifically designed to evaluate subject-level visual fidelity and identity-motion binding. Extensive experiments on MotionTwin-Bench and real-world videos demonstrate that WeLike2Party outperforms recent state-of-the-art methods in subject-level visual fidelity, identity-motion binding, and overall perceptual quality, particularly in multi-person interactions with substantial occlusion.

cs.CV↗