arXiv:2603.12630v1 Announce Type: cross Abstract: The rise of foundation models has driven the emergence of AI supply chains, where upstream foundation model providers offer fine-tuning and inference services to downstream firms developing domain-specific applications. Downstream firms pay providers to use their computing infrastructure to fine-tune models with proprietary data, creating a co-creation dynamic that enhances model quality. Amid concerns that foundation model providers and downstream firms may capture excessive consumer surplus, along with increasing regulatory measures, this study employs a game-theoretic model involving a provider and two competing downstream firms to analyze how policy interventions affect consumer surplus in the AI supply chain. Our analysis shows that policies promoting price competition in downstream markets (i.e., pro-price-competitive policies) boost consumer surplus only when compute or data preprocessing costs are high, while compute subsidies are effective only when these costs are low, suggesting these policies complement each other. In contrast, policies promoting quality competition in downstream markets (i.e., pro-quality-competitive policies) always improve consumer surplus. We also find that under pro-price-competitive policies or compute subsidies, both the provider and downstream firms can achieve higher profits along with greater consumer surplus, creating a win-win-win outcome. However, pro-quality-competitive policies increase the provider's profits while reducing those of downstream firms. Finally, as compute costs decline, pro-price-competitive policies may lose their effectiveness, whereas compute subsidies may shift from ineffective to effective. These findings offer insights for policymakers seeking to foster AI supply chains that are economically efficient and socially beneficial.

arXiv:2603.06664v1 Announce Type: cross Abstract: Diffusion Transformer (DiT)-based video generation models inherently suffer from bottlenecks in long video synthesis and real-time inference, which can be attributed to the use of full spatiotemporal attention. Specifically, this mechanism leads to explosive O(N^2) memory consumption and high first-frame latency. To address these issues, we implement system-level inference optimizations for a causal autoregressive video generation pipeline. We adapt the Self-Forcing causal autoregressive framework to sequence parallel inference and implement a sequence-parallel variant of the causal rotary position embedding which we refer to as Causal-RoPE SP. This adaptation enables localized computation and reduces cross-rank communication in sequence parallel execution. In addition, computation and communication pipelines are optimized through operator fusion and RoPE precomputation. Experiments conducted on an eight GPU A800 cluster show that the optimized system achieves comparable generation quality, sub-second first-frame latency, and near real-time inference speed. For generating five second 480P videos, a 1.58x speedup is achieved, thereby providing effective support for real-time interactive applications.

arXiv:2510.12264v2 Announce Type: replace Abstract: Active reasoning requires large language model (LLM) agents to interact with external sources and strategically gather information to solve problems in multiple turns. Central to this process is belief tracking: maintaining an accurate representation of the underlying state and uncertainty in understanding and solving the problem. However, due to limited reasoning capabilities, LLM-based agents often suffer belief deviation: their internal beliefs drift from the true problem state, leading to loss of state awareness and uninformative or repetitive actions. Once this happens, errors compound in the trajectories used for reinforcement learning (RL), leading to misattributed credits and limited exploration. To address this issue, we propose to track belief deviation and develop $\mathbf{T^3}$, a simple yet principled method that detects excessive deviation and truncates training trajectories to suppress uninformative tail effects. Hence, $\mathbf{T^3}$ preserves credits for informative prefixes and systematically improves policy optimization. Across 5 challenging tasks, $\mathbf{T^3}$ consistently enhances training stability and yields performance gains of up to 30 points while cutting token cost by up to 34%. These results highlight belief control as a key principle for building robust LLM agents capable of active reasoning.

arXiv:2504.10917v2 Announce Type: replace-cross Abstract: Recent advancements in large-scale pre-training have shown the potential to learn generalizable representations for downstream tasks. In the graph domain, however, capturing and transferring structural information across different graph domains remains challenging, primarily due to the inherent differences in graph topological patterns across various contexts. For example, a social network's structure is fundamentally different from that of a product co-purchase graph. Additionally, most existing models struggle to capture the rich topological complexity of graph structures, leading to inadequate exploration of the graph embedding space. To address these challenges, we propose GFSE, a universal pre-trained graph encoder designed to capture transferable structural patterns across diverse domains such as the web graph, social networks, and citation networks. GFSE is the first cross-domain graph structural encoder pre-trained with multiple self-supervised learning objectives. Built on a Graph Transformer, GFSE incorporates attention mechanisms informed by graph structural information, enabling it to encode intricate multi-level and fine-grained topological features within complex graph structures. The pre-trained GFSE produces generic and theoretically expressive positional and structural encoding for graphs, which can be seamlessly integrated with various downstream graph feature encoders, including graph neural networks for vectorized features and Large Language Models (LLMs) for text-attributed graphs. Comprehensive experiments on synthetic and real-world datasets demonstrate GFSE's capability to significantly enhance the model's performance while requiring substantially less task-specific fine-tuning.

arXiv:2602.15022v1 Announce Type: cross Abstract: Many generative tasks in chemistry and science involve distributions invariant to group symmetries (e.g., permutation and rotation). A common strategy enforces invariance and equivariance through architectural constraints such as equivariant denoisers and invariant priors. In this paper, we challenge this tradition through the alternative canonicalization perspective: first map each sample to an orbit representative with a canonical pose or order, train an unconstrained (non-equivariant) diffusion or flow model on the canonical slice, and finally recover the invariant distribution by sampling a random symmetry transform at generation time. Building on a formal quotient-space perspective, our work provides a comprehensive theory of canonical diffusion by proving: (i) the correctness, universality and superior expressivity of canonical generative models over invariant targets; (ii) canonicalization accelerates training by removing diffusion score complexity induced by group mixtures and reducing conditional variance in flow matching. We then show that aligned priors and optimal transport act complementarily with canonicalization and further improves training efficiency. We instantiate the framework for molecular graph generation under $S_n \times SE(3)$ symmetries. By leveraging geometric spectra-based canonicalization and mild positional encodings, canonical diffusion significantly outperforms equivariant baselines in 3D molecule generation tasks, with similar or even less computation. Moreover, with a novel architecture Canon, CanonFlow achieves state-of-the-art performance on the challenging GEOM-DRUG dataset, and the advantage remains large in few-step generation.

arXiv:2505.19645v4 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have achieved remarkable success across many applications, with Mixture of Experts (MoE) models demonstrating great potential. Compared to traditional dense models, MoEs achieve better performance with less computation. Speculative decoding (SD) is a widely used technique to accelerate LLM inference without accuracy loss, but it has been considered efficient only for dense models. In this work, we first demonstrate that, under medium batch sizes, MoE surprisingly benefits more from SD than dense models. Furthermore, as MoE becomes sparser -- the prevailing trend in MoE designs -- the batch size range where SD acceleration is expected to be effective becomes broader. To quantitatively understand tradeoffs involved in SD, we develop a reliable modeling based on theoretical analyses. While current SD research primarily focuses on improving acceptance rates of algorithms, changes in workload and model architecture can still lead to degraded SD acceleration even with high acceptance rates. To address this limitation, we introduce a new metric 'target efficiency' that characterizes these effects, thus helping researchers identify system bottlenecks and understand SD acceleration more comprehensively. For scenarios like private serving, this work unveils a new perspective to speed up MoE inference, where existing solutions struggle. Experiments on different GPUs show up to 2.29x speedup for Qwen2-57B-A14B at medium batch sizes and validate our theoretical predictions.

arXiv:2510.02356v3 Announce Type: replace-cross Abstract: The deployment of Large Language Models (LLMs) in embodied agents creates an urgent need to measure their privacy awareness in the physical world. Existing evaluation methods, however, are confined to natural language based scenarios. To bridge this gap, we introduce EAPrivacy, a comprehensive evaluation benchmark designed to quantify the physical-world privacy awareness of LLM-powered agents. EAPrivacy utilizes procedurally generated scenarios across four tiers to test an agent's ability to handle sensitive objects, adapt to changing environments, balance task execution with privacy constraints, and resolve conflicts with social norms. Our measurements reveal a critical deficit in current models. The top-performing model, Gemini 2.5 Pro, achieved only 59\% accuracy in scenarios involving changing physical environments. Furthermore, when a task was accompanied by a privacy request, models prioritized completion over the constraint in up to 86\% of cases. In high-stakes situations pitting privacy against critical social norms, leading models like GPT-4o and Claude-3.5-haiku disregarded the social norm over 15\% of the time. These findings, demonstrated by our benchmark, underscore a fundamental misalignment in LLMs regarding physically grounded privacy and establish the need for more robust, physically-aware alignment. Codes and datasets will be available at https://github.com/Graph-COM/EAPrivacy.

arXiv:2601.16905v2 Announce Type: replace-cross Abstract: Machine unlearning (MU) for large language models has become critical for AI safety, yet existing methods fail to generalize to Mixture-of-Experts (MoE) architectures. We identify that traditional unlearning methods exploit MoE's architectural vulnerability: they manipulate routers to redirect queries away from knowledgeable experts rather than erasing knowledge, causing a loss of model utility and superficial forgetting. We propose Geometric Routing Invariance Preservation (GRIP), an algorithm-agnostic framework for unlearning for MoE. Our core contribution is a geometric constraint, implemented by projecting router gradient updates into an expert-specific null-space. Crucially, this decouples routing stability from parameter rigidity: while discrete expert selections remain stable for retained knowledge, the continuous router parameters remain plastic within the null space, allowing the model to undergo necessary internal reconfiguration to satisfy unlearning objectives. This forces the unlearning optimization to erase knowledge directly from expert parameters rather than exploiting the superficial router manipulation shortcut. GRIP functions as an adapter, constraining router parameter updates without modifying the underlying unlearning algorithm. Extensive experiments on large-scale MoE models demonstrate that our adapter eliminates expert selection shift (achieving over 95% routing stability) across all tested unlearning methods while preserving their utility. By preventing existing algorithms from exploiting MoE model's router vulnerability, GRIP adapts existing unlearning research from dense architectures to MoEs.