每天知道
你的领域
出了什么

Multi-agent debate (MAD) is often used to improve large language model (LLM) reasoning, but sequential debate is rarely a neutral aggregator of agents' opinions. We show that sequential MAD suffers from a pronounced first-speaker bias: agents disproportionately shape the final answer when they speak first. As a result, placing a stronger model after weaker ones can substantially offset its reasoning advantage. We then focus on the disadvantaged strong-agent-last setting and ask whether personality prompting can mitigate this imbalance. Drawing on the Big Five model, we study agreeableness and extraversion as behavioral interventions applied to either the strong or weak side. We find that their effects are trait-specific. Influence consistently shifts in the direction of lower agreeableness, and assigning low agreeableness to the stronger agent helps restore its lost influence and improves final accuracy. Extraversion, by contrast, produces less systematic changes in influence and accuracy, with its clearest effect appearing in agents' verbosity. These findings show that effective MAD design depends not only on model capability, but also on how speaking order and induced interaction behavior shape the debate process.

LLM-based multi-agent systems coordinate specialized reasoning through aggregation, interaction, and adaptive control, yet their potential for graph learning remains unexplored. Graph learning is a natural setting for such systems because useful evidence may arise from heterogeneous local, long-range, global structural, and semantic perspectives whose relevance varies across instances. Existing LLM-based graph learning approaches primarily rely on single-agent reasoning, while multi-agent coordination has been studied mainly in general reasoning settings. Consequently, it remains unclear whether multiple specialized agents can improve graph learning and how coordination strategies should be designed and evaluated. To address this gap, we introduce GraphMAS, a systematic benchmark of multi-agent coordination for graph learning. GraphMAS builds a shared pool of graph reasoning specialists and organizes coordination along two dimensions, inter-agent interaction and runtime adaptivity, yielding four paradigms and seven representative coordination methods. Under a unified protocol, we evaluate these methods across seven text-attributed graphs, three domains, and two graph learning tasks. We find that heterogeneous graph perspectives are complementary, and that coordinating specialists improves over individual specialists and single-agent graph reasoning, with gains from decomposing reasoning across specialists rather than from broader evidence access alone. However, richer inter-agent interaction does not reliably help, whereas instance-adaptive specialist selection yields the strongest accuracy-efficiency trade-off. We further show that coordination can be learned over a fixed specialist pool and transfers to held-out graphs. GraphMAS therefore provides a controlled evaluation framework and empirical principles for understanding when and how multi-agent coordination benefits graph learning.

As multi-agent systems enter high-stakes domains, the possibility that agents may circumvent safety boundaries is a growing concern. Prior work has examined this risk primarily in adversarial settings, where agents are instructed or rewarded to communicate covertly and evade oversight. We show that benign agents can cross the same boundaries without adversarial incentives. We emulate a software-engineering workflow in which a planner represents a company hiring an external developer. The planner writes requirements and holds a company credential it is instructed not to disclose to the developer; a monitor screens their exchanges. Seven of nine tested frontier models disguise the credential in their requirements to help the developer recover it while evading the monitor, even after completing their assigned objective. For example, across 6,000 episodes with DeepSeek-V4-Pro, the planner attempts concealment in 16.9%; in 0.9%, the credential evades the monitor and is recovered and used by the developer. Despite the low per-episode rate, the risk compounds across repeated exchanges: at 0.9%, 105 independent episodes carry a 61.3% chance of at least one breach. Our analysis suggests that the planner misinterprets disguised disclosure as helping the developer and reads the nondisclosure rule as banning plaintext, not character codes or riddles. We call this behavior covert assistance. Monitoring favors concealment, while concealment gives way to direct disclosure when the recipient is presented as a human rather than an AI agent, so leakage persists. Giving the monitor the credential improves detection but requires trusting it with the secret. These risks, in models already used for software engineering, challenge oversight to distinguish authorized cooperation from task-advancing assistance that crosses safety boundaries.

Autonomous agents powered by large language models (LLMs) are rapidly evolving into an open agentic ecosystem. To support trustworthy collaboration, industry initiatives increasingly assess agent reputation from past behavior and provide performance leaderboards. However, reputation derived from past performance may not reliably predict an agent's behavior on new tasks, particularly when malicious agents can adapt their behavior and influence other agents during collaboration. We study reputation in multi-agent debate (MAD), where multiple agents answer the same query, debate to improve their answers, and aggregate them into a final output. We present MiniRep, a reputation-based aggregation system for MAD under malicious agents. To ground our threat model in established research, we construct an attack taxonomy drawing on reputation-system attacks and software-testing mutation operators, covering strategic exploitation of reputation and subtle corruption of agent proposals. Guided by this taxonomy, MiniRep evaluates agents based on both their behavior on the current task and their reputation over time, while preventing groups of agents with highly similar responses from dominating the final decision. We assess MiniRep across diverse tasks, LLM-agent compositions, corruption placements, and attack types drawn from our taxonomy. Our experimental results show that, MiniRep outperforms both conventional MAD aggregation and conventional reputation-based approaches on MATH no matter being attacked or not. Also, under a heterogeneous 10-agent setting on MATH, MiniRep outperforms all baselines in all 28 attack conditions.

Multi-agent debate (MAD) can improve large language model (LLM) reasoning by allowing multiple agents to exchange and critique their answers to the same task. However, the interactions that enable agents to correct mistakes can also spread adversarial errors and steer the agents toward an incorrect answer. Although some efforts have been made to examine particular attack types on MAD, systematic evaluation of MAD under diverse attacks remains limited. A central question is whether debate mitigates adversarial influence or amplifies it. In this paper, we present MADBench, a benchmark for evaluating the security of MAD. We organize attacks into a layered taxonomy following the MAD workflow, incorporating both established attacks and new strategies tailored to debate. We evaluate six attack families over 356 source tasks and 3,958 test cases, examining their effects on the final answer and the propagation of adversarial influence. Our results show that, under attacks, MAD does not necessarily improve LLM reasoning. Compared with a single-agent baseline, MAD can mitigate attacks on answer accuracy in question-answering tasks while amplifying unauthorized reads or writes in both question-answering and workspace tasks. Moreover, even when three out of five agents collude, the attack changes the final answer from correct to wrong on only 28.30% of tasks answered correctly without attack, while only 3.26% of initially correct honest agents switch to wrong answers during debate.

Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence extraction: objects, locations, and spatial relations are jointly constrained by the image and query, yet their dependencies do not imply an intrinsic left-to-right generation order. This distinction makes bidirectional diffusion a natural fit, allowing spatial hypotheses to emerge in parallel and be jointly refined through iterative denoising. We introduce GroundAnything, a 4B-parameter grounding foundation model that reconciles fast parallel decoding with precise localization through blockwise denoising. Training combines grounding pretraining from public datasets and dedicated data engines, direct AR-to-diffusion conversion with joint AR and diffusion objectives, supervised fine-tuning, and GRPO-based reinforcement post-training. Across 30 grounding benchmarks, our autoregressive variant, GroundAnything-VLM, establishes a new overall state of the art among similarly sized models at 72.42%, remaining competitive with GPT-6 Astra (71.35%). With entropy-guided decoding, GroundAnything also surpasses the prior state of the art at this scale, averaging 61.75% versus 53.32% for the fast MTP-based LocateAnything model. We further explore decoding strategies, showing that an optional self-speculative mode achieves a $4.51\times$ speedup over the AR counterpart with a 0.74 percentage-point drop in COCO F1mIoU. Infrastructure experiments show that progressive inference optimizations translate parallel decoding into practical speedups. These support efficient visual grounding in latency-sensitive real-world systems.

This paper demonstrate that whether masking-based token pruning helps or hurts worst-group robustness can be predicted before deployment, without labels or fine-tuning. A systematic study of semantic masking across 8 spurious-correlation benchmarks shows its effect on worst-group accuracy is highly unstable: it improves accuracy by up to 82.5% relative on some datasets and degrades it by up to 100% on others. We trace this instability to spurious inversion: background patches receive higher CLIP text-similarity than the true object when the spurious attribute is background-separable, inverting the assumption every text- and attention-guided pruning method relies on. We introduce the Spurious Inversion Metric (SIM), a label-free, pre-deployment diagnostic whose sign predicts this effect with statistical significance (binomial $p=0.035$) across all 8 datasets, and remains dependable across 6 CLIP architectures with a clean foreground/background split. Naive masking is itself a major source of risk: it causes the largest average-accuracy loss of any method we evaluate, and its own per-image segmentation step is a significant runtime bottleneck. To address this, we design a batched, synchronization-free GPU segmentation routine that cuts this overhead from 3.5$\times$ to 1.75$\times$ baseline. Gating deployment by SIM's sign recovers masking's benefits while avoiding its worst failures, matching or exceeding a strong pruning baseline on 7 of 8 datasets.

We present Multimodal Flow, a fully continuous generative model of language and vision. Most unified multimodal models either model both language and quantized images as discrete tokens or combine discrete language prediction with continuous image generation. The former introduces a visual quantization bottleneck. The latter requires modality-dependent objectives and sampling procedures. Fully continuous modeling avoids these trade-offs and enables a shared generative process, but remains underexplored for multimodal pretraining. Multimodal Flow introduces a unified continuous architecture that integrates multimodal continuous representations with a shared chunk-causal flow backbone. It organizes text blocks and images as ordered continuous hyperchunks, preserving textual token order and visual spatial structure. The backbone learns a single vector field over these hyperchunks through Flow Matching. Joint attention enables cross-modal interaction, while modality-specific feed-forward networks process each modality. The model predicts multiple target chunks in parallel during training and generates hyperchunks sequentially at inference. We instantiate MF-1 and pretrain it on multimodal data. Across 0.6B, 1.2B, and 1.6B scales, continued pretraining consistently improves multimodal modeling. With only 150B pretraining tokens, MF-1 achieves an average score of 82.8 across GenEval and DPG-Bench and 75.3 across VQAv2, MMBench, and POPE, remaining competitive with unified models trained on substantially more data. Under matched data, optimization, and parameter budgets, Multimodal Flow further outperforms representative hybrid and discrete models. These results establish continuous chunk-based embedding flow modeling as a new fully continuous paradigm for unified multimodal modeling. The related code and model are publicly released at https://github.com/hustvl/Multimodal-Flow.

Visual-token compression improves the efficiency of large vision-language models, but can expose failures that full-token evaluation misses. We study adversarial images that preserve full-token correctness yet induce errors after compression, even when both inference paths succeed on the clean image. Creating such failures is challenging because perturbing token importance can also damage the visual content needed for full-token inference. We propose Feature-Aware Token Attack (FATA), which couples attention suppression with cosine-based feature preservation on a fixed set of salient clean-image tokens. In the primary LLaVA-1.5-7B setting, FATA uses only vision-encoder gradients, without access to the deployed compressor, token budget, or downstream task. Across four visually dependent task subsets and four compressors under a controlled reconstruction protocol, FATA achieves SR = 96.3% full-token accuracy retention and CBR = 22.1% conditional blinding, compared with 89.8% and 15.7% for CAA. Ablations support the role of both objectives in balancing compressed-path failure against full-token preservation. FATA also has the lowest measured detection rate among four attacks across three evaluated detectors at a 5% false-positive rate. These findings motivate assessing adversarial robustness jointly across full-token and compressed inference.

Expert parallelism (EP) enables inference of large Mixture-of-Experts (MoE) models by placing their experts across multiple GPUs, but requires substantial communication between GPUs at every MoE layer. As contemporary MoE models activate more experts per token, this communication accounts for a growing fraction of inference time. The cost becomes particularly pronounced on PCIe-based consumer GPU systems, where all inter-GPU transfers traverse CPU memory. However, existing MoE-specialized EP communication libraries assume that direct GPU-to-GPU access is available, largely overlooking consumer GPUs. Therefore, most LLM frameworks instead rely on NCCL, whose CPU-staged communication incurs redundant PCIe transfers and competes with expert computation for GPU resources, limiting their overlap. We present ThunderEP, a novel communication design for such systems that removes the relay hops of traditional ring algorithm, moves data through DMA engines to avoid compute resource contention, and minimizes synchronization latency by reducing the polling overhead of completion flags in CPU memory. We integrate the proposed design into vLLM and evaluate it on three widely used MoE models. Experiments on two PCIe systems equipped with RTX 4090 and RTX 5090 GPUs show that ThunderEP achieves average speedups of 2.00$\times$ and 1.53$\times$ over NCCL for dispatch and combine, respectively, and up to 1.66$\times$ end-to-end speedup over state-of-the-art MoE inference frameworks.

A large audio language model (LALM) turns a minute of speech into 750-1,500 tokens and prefills every one. Image-token pruning often cuts after the language model's first layers, where image tokens draw little attention. Audio tokens draw much more attention there, and their ranking is still far from final, so audio needs a ranking before the language model runs. Surprisingly, the attention an audio token will receive across the language model is already linearly predictable from its encoder output, before the language model runs. A linear map, fitted in closed form without labels, predicts this all-layer attention ranking at $ρ\geq .69$ on eleven of thirteen LALMs. Our method, Triage, cuts audio tokens by this prediction and, on multiple choice, cuts again at layer 2, correcting the prediction with the attention observed there. Triage sets its compression without labels, under two budgets that limit how far its output may differ from the model's own full-audio output. At the conservative budget, its word error rate and accuracy stay within .04 of full audio. At the aggressive budget, Triage beats every baseline in all twelve transcription cases. On multiple choice, at 2.2-5x compression, it outperforms DART, the strongest baseline on average, by .043 in mean accuracy. Because it cuts before the language model, it raises the audio that fits in Qwen2.5-Omni-3B's context window from 21.8 to about 62 minutes. At its most compressive point, Triage lets one GPU serve 4x as many concurrent 5-minute streams of that model. Project page: https://audio-triage.github.io

Diffusion caching accelerates generation by replacing transformer computation with cached or predicted features at selected denoising steps. We introduce the Golden Path Hypothesis (GPH): under fixed inference conditions, prompt-independent cache schedules can achieve final-output quality comparable to the best prompt-specific schedules across prompts. We investigate the GPH across ten caching methods, four image and video models, and three cache ratios. Prompt-adaptive methods repeatedly select a small number of schedules, and reusing their most frequent schedules on new prompts closely matches the quality of prompt-specific choices. Exhaustive evaluation of 1.4 million schedules on four examples further identifies prompt-independent schedules that remain competitive on unseen prompts. To explain this transfer, we analyze denoising trajectories and the accumulation of caching errors. Latent-state trajectories exhibit similar structures across datasets and seeds, while an exact error decomposition shows that accumulated effects of earlier errors predict final latent-state error better than local approximation errors. This motivates searching for end-to-end schedules using final-output quality. With only a small set of examples, the resulting golden paths transfer across prompts and datasets, and can be tuned to the desired quality objective, including reconstruction fidelity or perceptual similarity.

Speculative decoding accelerates language model inference by verifying multiple draft tokens in a single target-model pass. Recent parallel drafters have achieved breakthrough performance in frontier production models, but their effectiveness deteriorates as the entropy of target distributions increases due to insufficient draft diversity. To overcome this bottleneck without sacrificing parallelism, we introduce UBTree, a parallel drafter that couples a Unigram proposer with a Bigram selector to construct drafting Trees. The unigram proposer is trained with the standard cross-entropy objective to generate candidate tokens independently for each position, while a lightweight bigram selector predicts transition scores between adjacent candidate pairs. Unlike the proposer, the selector is trained with a renormalized KL objective on high-temperature data. This tree-native training broadens the supervision beyond the greedy path, encouraging plausible alternative branches that improve the chance of accepting additional tokens during tree verification. Across seven standardized benchmarks with Qwen3-4B and Qwen3-8B, UBTree achieves an average speedup of $5.84$--$6.94\times$ over autoregressive decoding and outperforms DARTree in all 28 comparisons. Production-scale evaluation further demonstrates UBTree's advantage over frontier baselines such as DSpark.

As new evidence arrives, a sequence model must update what it remembers and how memory influences predictions. While Transformers incur computation and cache costs scaling with context length, fixed-state recurrent models offer constant-memory inference. However, linear and spectral recurrences traditionally rely on static transitions, failing to dynamically revise how stored representations decay or rotate. While recent selective architectures introduce input-dependent transitions, they assign independent controls to every memory mode, coupling control cost to state capacity. We show that high-dimensional spectral memory does not require high-dimensional control, and introduce Shared Phase and Retention Control for Efficient Adaptive Spectral Recurrence (SPARC). SPARC employs just two input-dependent scalar signals to coordinate memory retention and phase rotation across heterogeneous complex modes, while preserving mode-specific baseline timescales and frequencies. Its diagonal affine recurrence supports parallel associative scans for sequence-level BPTT as well as exact structured Real-Time Recurrent Learning (RTRL) for online credit assignment. Across partially observable continuous control, POPGym, and sequence classification, SPARC achieves a 9.09% relative return improvement on Walker-P and a 1.36% relative accuracy gain on FordA over second-best methods. On an NVIDIA Blackwell GPU, our implementation reduces recurrent-mixer training latency by 18.2%-34.2% in fixed-token workloads and accelerates scans by 3.1x-4.7x over an optimized RG-LRU baseline. These results show that two shared control signals can efficiently govern adaptive spectral memory across online and full-sequence settings. Code is available at https://github.com/Botwwt/sparc.

Progressive autoregressive image codecs provide an appealing paradigm for generative compression by quantizing continuous latents into discrete tokens, transmitting coarse-to-fine prefix tokens and generating the remaining suffix tokens at the decoder. However, their reconstruction quality is fundamentally limited by two residuals introduced along this pipeline: the quantization residual, arising from information loss during discrete tokenization, and the generation residual, resulting from imperfect autoregressive generation of the suffix tokens. To address these limitations, we introduce ResARC, a residual-aware autoregressive codec that explicitly compensates for both residuals at the decoder. Specifically, we generate the quantization residual with a diffusion transformer conditioned on the autoregressive decoding context, while requiring no additional side information. In parallel, we compute the generation residual at the encoder and employ a learned Generation Residual Codec to efficiently compress and transmit it for decoder-side correction. The recovered residuals are then integrated with the reconstructed latent representation and decoded through an adapted VAE decoder. Extensive experiments demonstrate that ResARC achieves competitive perceptual similarity while substantially improving distributional fidelity over leading generative codecs across the ultra-low bitrate regime. Code and models will be released soon.

Emerging non-matrix ML kernels, such as LayerNorm, GeLu, FFT or circular convolutions, demand low-latency, energy-efficient spatial accelerators beyond MatMul-centric arrays. STELLA presents a spatio-temporal elastic 16 nm coarse-grained reconfigurable array (CGRA) with a rapid configuration path, per-PE hardware loop control, and a low-latency, deeply pipelined elastic fabric with spatio-temporal data reuse. STELLA reaches up to 110 GOPS/mm2 at 850 MHz, and improves effective kernel throughput by 4.84-7.14x over baseline CGRAs.

When a camera revisits a previously observed region, a video world model should reproduce what was there before. This requires both remembering past observations and retrieving the right one for the current viewpoint. Key-value caches preserve visual detail but grow with video length; recurrent memory is compact but compresses history into a fixed-size state, so individual past observations are no longer directly accessible. We introduce LOCI, a hybrid spatial-memory architecture that keeps both representations. In half of the transformer blocks, main attention keeps a key-value cache of past observations; in the other half, it is restricted to the current chunk and complemented by a recurrent linear-attention memory whose reads and writes are conditioned on projective camera geometry, so viewpoint enters both memory addressing and stored content. Recurrent readouts flow into subsequent cache-backed blocks and supply their queries with accumulated scene context. On the public MIND memory benchmark and on held-out recorded trajectories, LOCI reproduces revisited content more faithfully than representative world models and a same-recipe full-softmax model; with full history, it lowers peak memory at equal length by about 30% relative to full softmax. With a bounded bank of retained observations, it streams long videos at constant memory and remains more faithful than full softmax under the same budget.

Fast matrix multiplication saves multiplications through exact cancellation, but rounding sums that mix token rows can leave contributions from later tokens in earlier language model outputs. This threatens prefix invariance, which multiple-choice likelihood scoring relies on: a scored likelihood must depend only on its allowed prefix. On Qwen2.5-14B-Instruct, two fast FP8 realizations repaired to ordinary-looking accuracy still change the answers chosen by likelihood on 5.83% and 10.00% of 240 OpenBookQA items when only the text after the allowed prefix is replaced with the bf16 model's own greedy continuation. Both row-local controls, the bf16 model and a deployed FP8 matrix multiplication kernel, change none. Accuracy thus does not certify prefix invariance, and the stability criteria we analyze cannot tell realizations apart: across all 512 sign variants of two-level Strassen they stay constant while teacher-forced perplexities span a 772.4$\times$ range on the same model. We therefore construct certified realizations of two-level Strassen on bounded integer codes that quantize token rows independently, then mix and cancel exactly before rescaling, using 49 block multiplications instead of 64. Our certificate guarantees bitwise equality to a prescribed row-local classical int8 operator at the same quantization specification, so every certified realization inherits its prefix invariance. Certification thus turns realization choice into a pure cost decision: which certified realization runs can no longer change a single scored likelihood.

Low-precision training rounds tensors that the backward pass reads again, often for several gradients; each use can read the forward's rounded value, the original, or a new random rounding. This backward-state policy looks like a memory and precision detail, settled by copy accuracy and final loss. We argue that it is part of the learning algorithm, and that neither check shows whether it is right. Copy accuracy does not decide the outcome: in three pairs of 390M runs with an emulated FP8 backward, training fails when attention's backward reuses the forward's rounded output and succeeds with a new rounding from the same distribution. Even the most accurate copy, the original itself, can be wrong by our reference: the gradient of the forward pass as it actually ran, with gradients passed through rounding unchanged. For example, a normalization output stored in low precision feeds two gradients: the gain's gradient needs the original, but the next layer's weight gradient needs the rounded value that layer multiplied. Final loss, the other check, does not rule out the error of reading the original for both: it persists in models trained with such a store, while planned loss comparisons stay within a margin fixed in advance. We therefore derive from this reference which value each use must read, or which substitute gives the same gradient on average with the forward held fixed, and check these per-use requirements on single operators, without training. In three tests using PyTorch and Transformer Engine, the requirements predicted beforehand whether reuse changes what the backward computes on average relative to an independent copy, and every prediction held. Backward-state policy is thus part of the learning algorithm: it should be specified and checked use by use, not settled by copy accuracy and final loss.

Scaling attention parameters can improve language model quality, but retaining full token histories makes additional heads costly at long contexts. Furthermore, since attention retrieves and combines contextual information, parameter scaling should also support longer contexts. We therefore ask whether attention parameter scaling can directly enable efficient and effective context scaling. We introduce NAMOH, an architecture-native sparse attention mechanism that activates $K$ of $H$ heads per token. Each head retains only its assigned tokens and performs causal attention within this subsequence. Head selection thus jointly determines active parameters and available context without scanning the full history. Under balanced assignments, increasing $H$ at fixed $K$ shortens head histories and reduces per-token key-value (KV) access without increasing total KV storage. We further support head-relative rotary position embeddings to shorten positional spans within routed subsequences, aiming to mitigate position-induced attention noise. Experiments show that NAMOH can outperform fully activated models with the same total parameters, while enabling more efficient long-context inference than smaller dense models with matched active parameter counts. It remains compatible with GQA and existing sparse attention mechanisms. We hope this work offers a new path for scaling attention, with parameter scaling directly enabling context scaling.

World-action models (WAMs) leverage pretrained video models to improve generalization in robot control by jointly predicting future visual states and actions. This capability comes at a substantial inference cost, as dense future-frame tokens are repeatedly processed during denoising. Prior methods address this by token pruning that prioritizes visual fidelity to reduce denoising costs in video diffusion models. However, these methods do not use action relevance to determine which future-frame tokens to retain during joint denoising in WAMs. In this paper, we propose Sparse-WAM, a training-free framework for action-guided sparse imagination that selectively processes future-frame tokens to accelerate WAM inference. We observe substantial overlap in the spatial distribution of attention from action tokens to future-frame tokens (action-to-future attention) between consecutive denoising steps, despite continued updates to the future representations. Motivated by this, we develop Action-Guided Token Selection to retain frame-specific action-relevant regions together with cross-frame context. However, a naive implementation can incur attention-scoring and token-packing overhead that offsets the computational savings from pruning. We therefore introduce Pilot, an efficient engine that reduces sparse inference overhead through lightweight scoring and cross-step reuse of token selections. On LIBERO with FastWAM-Joint and RoboLab-120 with Cosmos 3 Edge, Sparse-WAM achieves inference speedups of approximately $2.0\times$ and $1.8\times$, respectively, over dense eager inference on an NVIDIA RTX 4090, while largely preserving task performance.

Multimodal mixture-of-experts (MoE) models combine sparse expert activation with visual-language capabilities, yet their inference remains costly because long visual-token sequences repeatedly incur attention, routing, dispatch, and expert-MLP computation. Existing methods typically compress either the token or expert dimension, leaving redundancy along the other. Our analysis reveals two complementary regularities: the depth required for visual propagation varies across inputs, while text-token routing exhibits concentrated and recurrent expert-importance patterns. Based on these observations, we propose DecoMoE, a two-dimensional structured compression framework that decouples visual propagation from expert computation. The Sample-Adaptive Visual Boundary (SAVB) predicts an input-dependent visual-exit layer at which the visual-token block is removed. The Routing-Calibrated Expert Prefix (RCEP) reorders experts offline using text-token routed mass and, from this predicted exit layer onward, retains at each MoE layer the shortest contiguous prefix covering a target routed-mass fraction. We evaluate DecoMoE on Qwen3-VL-MoE and InternVL3.5-30B-A3B across six benchmarks. On Qwen3-VL-MoE, DecoMoE retains 97.91% of dense-baseline performance while reducing computation from 27.06 to 16.73 TFLOPs and latency from 0.44 to 0.26 seconds, yielding a 1.69x speedup. Code will be available at https://github.com/ShawnTan86/DecoMoE.

Diffusion Transformers (DiTs) have become a dominant architecture for video generation, but their efficiency is limited by the quadratic complexity of full attention. Sparse attention reduces this cost by retrieving important blocks and computing attention only within them, but inaccurate retrieval can either degrade generation quality or yield unnecessary computation. We identify two retrieval mismatches in methods that retrieve blocks using the averaged representations of query and key blocks: (i) query-side aggregation mismatch, where averaging queries before Softmax fails to preserve their individual attention preferences, and (ii) key-side clustering metric mismatch, where standard Euclidean clustering in the original key space can group keys with dissimilar QK scores under the current query, so their average representation may not accurately represent how the current query scores individual keys. These mismatches can lead to inaccurate block retrieval. To address these mismatches, we propose PARK, a training-free sparse attention method for accurate block retrieval. PARK retains every original query, independently normalizes its attention over key blocks, and then averages these distributions within each query block. It also uses information from the current queries to transform keys before clustering, so that keys receiving similar QK scores are grouped together. A fused GPU kernel further reduces the overhead of block retrieval. Experiments on HunyuanVideo and Wan demonstrate that PARK improves block retrieval accuracy and preserves generation quality while accelerating inference, achieving the best quality-efficiency trade-off among the compared sparse attention methods.

Test-time scaling has recently emerged as a powerful approach for improving LLM reasoning by allocating additional computation during inference, substantially enhancing accuracy on challenging tasks such as mathematics and coding. To accelerate the exploration of reasoning paths, recent studies proposed speculative execution. However, we show that supporting speculative execution poses two unique challenges for LLM serving systems: (1) an explosion in the search space of candidate paths and (2) frequent, fine-grained verification tasks for candidates. To address these challenges, this paper proposes SpecScale, a serving system for efficient speculative execution. We introduce three techniques to reconcile the trade-off between latency and computational overhead: (1) early pruning of low-quality candidate paths, (2) deduplicating computation across redundant candidate paths, and (3) deferring fine-grained verification tasks. We evaluate SpecScale on challenging reasoning benchmarks, including MATH and Olympiad. Our results show that SpecScale significantly outperforms both non-speculative and recent speculative approaches, delivering substantial improvements in throughput and latency while preserving answer quality.

Vision-language models face a fundamental scaling bottleneck: the number of visual tokens grows with both temporal duration and spatial resolution, making long-video understanding expensive for the vision encoder and the language model. Existing methods often compress visual tokens after dense encoding, creating a mismatch between the representation used during training and the compact interface required at deployment. We present CoVisco, a codec-native vision encoder with native token compression for unified image-video understanding. By combining codec-native input support with segmented attention, CoVisco can encode long visual inputs in a single forward pass without forming dense patch-to-patch interactions across all frames. Each temporal segment is equipped with learnable abstract tokens that learn a compact segment-level representation, while fine-grained patch tokens remain available throughout the encoder. Alternating intra-segment and abstract-communication layers preserve video-level context through the abstract-token channel. A lightweight selector further exposes either abstract tokens alone or abstract tokens augmented with a runtime-selected subset of patch tokens, yielding a compact visual interface that reduces the visual context and prefill burden of downstream MLLMs while retaining fine-grained evidence when needed. Pretrained with contrastive objectives on 565M image--text pairs and 6.4M videos, CoVisco shows competitive performance on video-oriented embedding and multimodal understanding benchmarks. In the evaluated four-segment, 64-frame setting, abstract-only inference uses only 400 visual tokens while achieving video-understanding performance close to, and on some benchmarks exceeding, OneVision-Encoder. Selected patch tokens further improve fine-grained video reasoning. Project URL: https://github.com/ernie-research/CoVisco.git

A transformer language model performs a bounded amount of computation per token, and recent work by Vishal Sikka, former CEO of Infosys, argues that this bound limits which tasks a model can carry out or verify (arXiv:2507.07505). We ask how much of the budget beneath that ceiling is spent on work the model has already done. Serving is stateless across requests: a model that answers a second question about a document recomputes the document's attention state from the first token. On seven real-world datasets, 98.7% of prompt tokens were text the model had already read. We present Galahad, a memory layer for vLLM, SGLang and llama.cpp that makes this reading a one-time cost. Taliesin saves the model's key-value (KV) state for a block of text and loads it on the next request that contains the same bytes, instead of recomputing it. Blaise keeps the documents themselves and passes the model only the section a question needs. On a recall test with 100 facts hidden in a 97,000-token corpus (Gemma 4 31B), Taliesin alone let the model attend to the whole corpus and answered 98 of 100 on llama.cpp at 3.0 s and 572 J per question, against 10 of 100, 9.3 s and 2,754 J for the same model without Galahad, which could hold only the last 12,000 tokens. With Blaise added, the model read about 668 tokens per question and answered 100 of 100 on all three runtimes at 0.59-0.64 s and 200-213 J; a tuned RAGFlow pipeline answered 77. Storing the corpus is a one-time cost of about 100 s and 28 kJ, whose energy is recovered after 13 questions. Restored state is bit-identical: all 262,144 output logits matched after restart, rehydration and hot-load. Galahad worked with all 30 models we tested under vLLM, and it fails closed: any load that does not pass its checks is recomputed. Together these results move LLM serving from stateless to stateful inference.

World-action models (WAMs) jointly generate future world states and actions through iterative denoising, using shared weights to process heterogeneous semantic streams of video, proprioceptive, and action tokens. Quantization reduces inference cost, but comparable numerical errors in different streams can have markedly different effects on final actions, making numerical accuracy alone insufficient for reliable control. We introduce SteerQuant, a 4-bit quantization framework for WAMs that steers errors toward computations with less influence on final actions. It maps how each stream's quantization errors affect final actions and uses this map to guide shared channel scaling. Activation scaling is further calibrated for each stream and denoising step to accommodate changes in activation ranges and action impact. This adapts quantization to different stream requirements without duplicating weights or increasing bit-widths for selected streams. To reduce the extra kernel launches and memory traffic introduced by scaling, we develop Rudder, a 4-bit inference engine for WAMs that fuses scaling and output compensation into low-bit kernels. Under W4A8 and W4A4, SteerQuant maintains mean LIBERO success within 0.8 percentage points of full precision, while delivering up to $2.23\times$ denoising speedup over BF16 across three WAMs with reduced peak GPU memory usage. On a real dual-arm robot, W4A8 deployment achieves a $1.35\times$ end-to-end inference speedup while maintaining average task success relative to BF16.

Large language model (LLM) inference is increasingly moving toward lower precision to realize the throughput of hardware accelerators, but aggressive post-training quantization (PTQ) can degrade model quality. We present QATFactory, an open-source framework for deployment-aligned quantization-aware distillation (QAD) and reinforcement learning (QARL). QATFactory simulates deployment-time quantization while performing matrix multiplications in BF16, allowing models to adapt to quantization noise without requiring training hardware that natively supports the target format; for example, it supports NVFP4 training on H100 GPUs, which lack FP4 Tensor Cores. The framework supports NVFP4, MXFP4, and $\text{llama}.\text{cpp}$'s Q4_K format; dense and mixture-of-experts models; and both full-parameter and LoRA-based training. It exports checkpoints directly to vLLM and $\text{llama}.\text{cpp}$ without an additional lossy conversion step or added inference overhead. With QATFactory, we conduct extensive experiments on models ranging from 8B to 230B parameters and evaluate exported checkpoints in production inference engines. Across models and formats, QAD consistently improves deployed-model quality over strong PTQ baselines. On Qwen3.5-9B, QAD achieves average benchmark accuracies of 68.9% under NVFP4 and 66.0% under MXFP4, outperforming the best PTQ results of 65.4% and 56.4%, respectively. Through our experiments, we found that although both FP4 formats quantize weights and activations at deployment, the best training strategy is format-dependent: NVFP4 generally performs better when only weights are quantized during training, whereas MXFP4 benefits from quantizing both weights and activations. At a fixed training token budget, training on fewer 32K sequences improves average accuracy by 1.9 points over training on more 4K sequences.

Modern transformers pair impressive capabilities with substantial memory and compute demands. Low-rank weight factorization reduces both while keeping the matrices dense, and thus efficient on standard hardware. Existing methods, however, choose the subspace to remove from each weight matrix with local closed-form criteria: activation energy, layer-wise reconstruction error, or a quadratic approximation of the loss. These criteria ignore how errors propagate through the network, so at high compression the errors compound with depth and performance collapses. We introduce Learnable Subspace Projections (LSP), which instead learns the subspaces to discard end-to-end. Each linear layer, or tied group of layers that read the same activations, is assigned an orthogonal projector. All projectors are optimized jointly against a global objective--the KL divergence to the dense model's output distribution or the model's original training loss--while the pretrained weights remain frozen. Projectors are initialized from a whitened SVD truncation, and ranks are allocated by the output KL each projector induces per parameter saved. After training, the projectors merge into standard low-rank factors, with each tied group sharing one factor. In attention, this also lets the model cache one narrow latent in place of full keys and values. Across LLMs (OPT-125M/1.3B, Qwen3-4B, Llama-2-7B) and ViT-B/16, LSP outperforms baselines, and its advantage widens as compression increases. At -70% compression, LSP brings Llama-2-7B to 10.9 WikiText-2 perplexity and 42.2% mean zero-shot accuracy, versus 13.3 and 36.0% for the strongest baseline. The factorized model decodes up to 1.6x faster than the dense model at small batch sizes, and aching the shared latent shrinks the combined memory of weights and KV cache by 13.5x at a 128k-token context, versus at most 6.5x for untied baseline factorizations.

When inference demand exceeds available compute capacity, model providers must decide which requests should be served first. Users have different tolerances for delay from an LLM API, but current priority pricing schemes compress these differences into coarse fixed-price service tiers. We design an inference auction that allows users to bid for faster service. Our auction allocates priority in an economically efficient way without sacrificing latency, and we develop fast algorithms for implementing prices that incentivize truthful bidding. We also design an autobidding agent for our inference auction, where users specify an inference budget and the autobidder dynamically adjusts its bids over time to maximize user utility subject to the budget constraint. Experiments validate the practicality of our auction: it increases system welfare while maintaining the cache utilization and latency advantages of SGLang, a state-of-the-art inference serving framework.