arXiv subjects

Junxiong Wang

Publications and source records attributed to Junxiong Wang.

At least 19 recordsLinked to original sources

arXiv2026

When RL Meets Adaptive Speculative Training: A Unified Training-Serving System

Speculative decoding can significantly accelerate LLM serving, yet most deployments today disentangle speculator training from serving, treating speculator training as a standalone offline modeling problem. We show that this decoupled formulation introduces substantial deployment and adaptation lag: (1) high time-to-serve, since a speculator must be trained offline for a considerable period before deployment; (2) delayed utility feedback, since the true end-to-end decoding speedup is only known after training and cannot be inferred reliably from acceptance rate alone due to model-architecture and system-level overheads; and (3) domain-drift degradation, as the target model is repurposed to new domains and the speculator becomes stale and less effective. To address these issues, we present Aurora, a unified training-serving system that closes the loop by continuously learning a speculator directly from live inference traces. Aurora reframes online speculator learning as an asynchronous reinforcement-learning problem: accepted tokens provide positive feedback, while rejected speculator proposals provide implicit negative feedback that we exploit to improve sample efficiency. Our design integrates an SGLang-based inference server with an asynchronous training server, enabling hot-swapped speculator updates without service interruption. Crucially, Aurora supports day-0 deployment: a speculator can be served immediately and rapidly adapted to live traffic, improving system performance while providing immediate utility feedback. Across experiments, Aurora achieves a 1.5x day-0 speedup on recently released frontier models (e.g., MiniMax M2.1 229B and Qwen3-Coder-Next 80B). Aurora also adapts effectively to distribution shifts in user traffic, delivering an additional 1.25x speedup over a well-trained but static speculator on widely used models (e.g., Qwen3 and Llama3).

cs.LG↗

arXiv2026

ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System

Large language models(LLMs) are now used to power complex multi-turn agentic workflows. Existing systems run agentic inference by loosely assembling isolated components: an LLM inference engine (e.g., vLLM) and a tool orchestrator (e.g., Kubernetes). Although agentic workflows involve multiple LLM and tool requests, these systems schedule and allocate resources separately on a per-request basis, without end-to-end knowledge of the workflow. This leads to sub-optimal management of KV cache and tool execution environments. To address the challenges, we propose ThunderAgent, a fast, simple, and program-aware agentic inference system. We first abstract agentic workflows as LLM Programs, enabling a unified view of heterogeneous resources, including KV caches, system states, and external tool assets such as disk memory and network ports. Built upon this abstraction, ThunderAgent introduces a program-aware scheduler and a tool resource manager designed to maximize KV cache hit rates, mitigate memory imbalances, and enable asynchronous environment preparation. Evaluations across coding, routing, and scientific discovery agents demonstrate that ThunderAgent achieves 1.5-3.6x throughput improvements in serving, 1.8-3.9x in RL rollout, and up to 4.2x disk memory savings compared to state-of-the-art inference systems. To facilitate reproducibility and support future development, we open-source the system implementations of the whole ThunderAgent at: https://github.com/Agentic-Kinetics/ThunderAgent.

cs.OS↗

arXiv2026

Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation

Hybrid linear attention models offer an appealing path to faster long-context inference: they reduce the quadratic cost and KV-cache burden of full softmax attention while retaining much of the quality of Transformer models. A practical way to obtain such models is to convert a pretrained Transformer instead of pretraining a new architecture from scratch, but this conversion is still brittle. Simply copying the teacher attention projections into a Gated DeltaNet (GDN) student does not specify the new recurrent decay, write, and output-gating dynamics. As a result, the converted model often starts in a poor dynamical regime and must spend many distillation tokens repairing initialization rather than learning the remaining teacher behavior. We propose Taylor-Calibrate, a lightweight initialization method for hybrid GDN students. The method uses Taylor-guided teacher attention statistics to set the value projection, memory timescale, write gates, and output gate, then applies a short per-layer alignment step to match each converted layer to the teacher output. Across four teacher settings and three retained-layer policies, Taylor-Calibrate gives substantially stronger zero-shot students, with up to an 88x improvement in a representative ablation, and reaches matched recovery targets with 4.9x--9.2x fewer training tokens than naive conversion.

cs.LG↗

arXiv2026

Understanding and Improving Length Generalization in Hierarchical Sparse Attention Models

Effectively processing long contexts is a critical challenge for language models. While standard Transformers are limited by quadratic complexity and poor length extrapolation, alternative architectures like sliding window attention and state space models sacrifice the ability to effectively utilize the full context due to their fixed-size memory. Chunk-based sparse attention has emerged as a promising paradigm for extreme length generalization, yet the key architectural principles underpinning its success are not yet fully understood. In this work, we present a systematic dissection of these models to identify the core components driving their performance. Through a unified framework and comprehensive ablation studies, we demonstrate that a combination of three design principles is critical: (1) an expressive, non-linear Chunk Encoder with a dedicated CLS token to produce representations for retrieval; (2) a Bypassing Residual Path to stably integrate retrieved global information without it being overridden by the local residual stream; and (3) enforced selection sparsity during pre-training to bridge the train-test distribution gap. We provide a theoretical motivation for intra-chunk information processing and landmark generation. By combining these principles, we establish a new state-of-the-art for training-free length extrapolation, successfully generalizing models trained on a 4K context to 32 million tokens on RULER and BABILong. Our findings provide a clear and empirically-grounded set of design principles for developing future, highly-capable long-context language models.

cs.CL↗

arXiv2026

Introspective Diffusion Language Models

Diffusion language models promise parallel generation, yet still lag behind autoregressive (AR) models in quality. We stem this gap to a failure of introspective consistency: AR models agree with their own generations, while DLMs often do not. We define the introspective acceptance rate, which measures whether a model accepts its previously generated tokens. This reveals why AR training has a structural advantage: causal masking and logit shifting implicitly enforce introspective consistency. Motivated by this observation, we introduce Introspective Diffusion Language Model (I-DLM), a paradigm that retains diffusion-style parallel decoding while inheriting the introspective consistency of AR training. I-DLM uses a novel introspective strided decoding (ISD) algorithm, which enables the model to verify previously generated tokens while advancing new ones in the same forward pass. From a systems standpoint, we build I-DLM inference engine on AR-inherited optimizations and further customize it with a stationary-batch scheduler. To the best of our knowledge, I-DLM is the first DLM to match the quality of its same-scale AR counterpart while outperforming prior DLMs in both model quality and practical serving efficiency across 15 benchmarks. It reaches 69.6 on AIME-24 and 45.7 on LiveCodeBench-v6, exceeding LLaDA-2.1-mini (16B) by more than 26 and 15 points, respectively. Beyond quality, I-DLM is designed for the growing demand of large-concurrency serving, delivering about 3x higher throughput than prior state-of-the-art DLMs.

cs.AI↗

arXiv2026

Squeeze Evolve: Unified Multi-Model Orchestration for Verifier-Free Evolution

We show that verifier-free evolution is bottlenecked by both diversity and efficiency: without external correction, repeated evolution accelerates collapse toward narrow modes, while the uniform use of a high-cost model wastes compute and quickly becomes economically impractical. We introduce Squeeze Evolve, a unified multi-model orchestration framework for verifier-free evolutionary inference. Our approach is guided by a simple principle: allocate model capability where it has the highest marginal utility. Stronger models are reserved for high-impact stages, while cheaper models handle the other stages at much lower costs. This principle addresses diversity and cost-efficiency jointly while remaining lightweight. Squeeze Evolve naturally supports open-source, closed-source, and mixed-model deployments. Across AIME 2025, HMMT 2025, LiveCodeBench V6, GPQA-Diamond, ARC-AGI-V2, and multimodal vision benchmarks, such as MMMU-Pro and BabyVision, Squeeze Evolve consistently improves the cost-capability frontier over single-model evolution and achieves new state-of-the-art results on several tasks. Empirically, Squeeze Evolve reduces API cost by up to $\sim$3$\times$ and increases fixed-budget serving throughput by up to $\sim$10$\times$. Moreover, on discovery tasks, Squeeze Evolve is the first verifier-free evolutionary method to match, and in some cases exceed, the performance of verifier-based evolutionary methods.

cs.AI↗

arXiv2026

CARE: Covariance-Aware and Rank-Enhanced Decomposition for Enabling Multi-Head Latent Attention

Converting pretrained attention modules such as grouped-query attention (GQA) into multi-head latent attention (MLA) can improve expressivity without increasing KV-cache cost, making it attractive for efficient inference. However, many practical conversion baselines rely on weight-only low-rank approximations (e.g., SVD-style initializations) and uniform rank allocation. They focus on minimizing the difference between weight matrices rather than on how those weights affect input activations, ignore the covariance structure of activations, and enforce uniform rank across layers, causing activation drift and degraded attention fidelity. To address these issues, we propose CARE, a Covariance-Aware, Rank-Enhanced MLA conversion pipeline under a fixed KV width. CARE introduces three key steps: (i) activation-preserving factorization, which aligns the approximation with the actual input activations rather than just the weights; (ii) adjusted-rank allocation, which spreads a fixed KV budget across layers by giving more capacity to layers that need it most; and (iii) KV-parity mapping, which reparameterizes the converted K and V to fit the MLA format while keeping the KV-cache size unchanged. Our method outperforms a uniform-rank SVD baseline on Qwen3-4B/30B-A3B-Instruct-2507 and Llama-3.1-8B/70B-Instruct, reducing one-shot perplexity by up to 215x and improving mean accuracy by up to 1.70x at matched KV budgets. With a brief post-SVD healing fine-tune, we fully recover the original model's accuracy.

cs.LG↗

arXiv2026

$V_1$: Unifying Generation and Self-Verification for Parallel Reasoners

Test-time scaling for complex reasoning tasks shows that leveraging inference-time compute, by methods such as independently sampling and aggregating multiple solutions, results in significantly better task outcomes. However, a critical bottleneck is verification: sampling is only effective if correct solutions can be reliably identified among candidates. While existing approaches typically evaluate candidates independently via scalar scoring, we demonstrate that models are substantially stronger at pairwise self-verification. Leveraging this insight, we introduce $V_1$, a framework that unifies generation and verification through efficient pairwise ranking. $V_1$ comprises two components: $V_1$-Infer, an uncertainty-guided algorithm using a tournament-based ranking that dynamically allocates self-verification compute to candidate pairs whose relative correctness is most uncertain; and $V_1$-PairRL, an RL framework that jointly trains a single model as both generator and pairwise self-verifier, ensuring the verifier adapts to the generator's evolving distribution. On code generation (LiveCodeBench, CodeContests, SWE-Bench) and math reasoning (AIME, HMMT) benchmarks, $V_1$-Infer improves Pass@1 by up to $10%$ over pointwise verification and outperforms recent test-time scaling methods while being significantly more efficient. Furthermore, $V_1$-PairRL achieves $7$--$9%$ test-time scaling gains over standard RL and pointwise joint training, and improves base Pass@1 by up to 8.7% over standard RL in a code-generation setting.

cs.CL↗

arXiv2026

Fairness-in-the-Workflow: How Machine Learning Practitioners at Big Tech Companies Approach Fairness in Recommender Systems

Recommender systems (RS), which are widely deployed across high-stakes domains, are susceptible to biases that can cause large-scale societal impacts. Researchers have proposed methods to measure and mitigate such biases - but translating academic theory into practice is inherently challenging. Through a semi-structured interview study (N=11), we map the RS practitioner workflow within large technology companies, focusing on how technical teams consider fairness internally and in collaboration with legal, data, and fairness teams. We identify key challenges to incorporating fairness into existing RS workflows: defining fairness in RS contexts, balancing multi-stakeholder interests, and navigating dynamic environments. We also identify key organization-wide challenges: making time for fairness work and facilitating cross-team communication. Finally, we offer actionable recommendations for the RS community, including practitioners and HCI researchers.

cs.HC↗

arXiv2026

Distilling Token-Trained Models into Byte-Level Models

Byte Language Models (BLMs) have emerged as a promising direction for scaling language models beyond tokenization. However, existing BLMs typically require training from scratch on trillions of bytes, making them prohibitively expensive. In this paper, we propose an efficient distillation recipe that converts existing token-trained LLMs into BLMs while retaining comparable capabilities. Our recipe follows a two-stage curriculum: (1) Progressive Knowledge Distillation, which aligns byte-level representations with the embeddings of the token-trained teacher model; and (2) Byte-Level Supervised Fine-Tuning, which enables end-to-end generation entirely in the byte space. We validate our approach across multiple model families, including Llama, Qwen, and OLMo, and demonstrate that the distilled BLMs retain most of the teacher models' performance using only approximately 125B bytes.

cs.CL↗

arXiv2025

Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost

The KV cache is a dominant memory bottleneck for LLM inference. While 4-bit KV quantization preserves accuracy, 2-bit often degrades it, especially on long-context reasoning. We close this gap via an algorithm-system co-design for mixed-precision KV caching: Kitty. On the algorithm side, extensive experiments show that Dynamic Channel-wise Precision Boost -- which ranks Key-cache channels by sensitivity and keeps only a small fraction at higher precision -- maintains near-zero loss in accuracy drop while approaching 2-bit memory. The main challenge is handling dynamic 4-bit channel boosts while keeping the page layout coalesced and the dequantization uniform, with no scattered reads or hard-coded masks. Kitty addresses these issues by decompose each mixed-precision Key page into two tensors with unified 2-bit precision. Based on this, Kitty provides a page-centric KV layout, Triton-compatible page dequantization kernels, and a lightweight runtime pipeline that preserves coalescing and avoids divergence. Across seven tasks and two model families (Qwen3, LLaMA3), Kitty cuts KV memory by nearly 8x with negligible accuracy loss, enabling up to 8x larger batches and 2.1x-4.1x higher throughput under the same memory budget. We release the full implementation of Kitty at https://github.com/Summer-Summer/Kitty.

cs.LG↗

arXiv2025

Beat the long tail: Distribution-Aware Speculative Decoding for RL Training

Reinforcement learning(RL) post-training has become essential for aligning large language models (LLMs), yet its efficiency is increasingly constrained by the rollout phase, where long trajectories are generated token by token. We identify a major bottleneck:the long-tail distribution of rollout lengths, where a small fraction of long generations dominates wall clock time and a complementary opportunity; the availability of historical rollouts that reveal stable prompt level patterns across training epochs. Motivated by these observations, we propose DAS, a Distribution Aware Speculative decoding framework that accelerates RL rollouts without altering model outputs. DAS integrates two key ideas: an adaptive, nonparametric drafter built from recent rollouts using an incrementally maintained suffix tree, and a length aware speculation policy that allocates more aggressive draft budgets to long trajectories that dominate makespan. This design exploits rollout history to sustain acceptance while balancing base and token level costs during decoding. Experiments on math and code reasoning tasks show that DAS reduces rollout time up to 50% while preserving identical training curves, demonstrating that distribution-aware speculative decoding can significantly accelerate RL post training without compromising learning quality.

cs.LG↗

arXiv2025

Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining

An increasing number of LLMs employ Mixture-of-Experts (MoE) architectures where the feed-forward layer is replaced by a pool of experts and each token only activates a small subset of them. During autoregressive generation, these models often enter a memory-bound regime even for moderate batch sizes because the average expert load grows more slowly than in an equivalent dense feedforward layer. Consequently, MoE latency is governed by the number of activated experts. We introduce a framework for dynamically re-routing token-to-expert mapping to lower this number (and thus, the decode latency) while preserving a comparable quality. Our best results use a batch-aware routing that works by having tokens piggyback experts that have already been loaded into memory due to being crucial to other tokens within the same batch. Empirically, we evaluate our method on the Qwen3-30B and Qwen3-235B models with a batch size of $16$. Without any statistically significant loss in accuracy, our approach achieves latency reductions of $39\%$ and $15\%$ in the MoE layer decode latency, respectively.

cs.LG↗

arXiv2025

M1: Towards Scalable Test-Time Compute with Mamba Reasoning Models

Effective reasoning is crucial to solving complex mathematical problems. Recent large language models (LLMs) have boosted performance by scaling test-time computation through long chain-of-thought reasoning. However, transformer-based models are inherently limited in extending context length due to their quadratic computational complexity and linear memory requirements. In this paper, we introduce a novel hybrid linear RNN reasoning model, M1, built on the Mamba architecture, which allows memory-efficient inference. Our approach leverages a distillation process from existing reasoning models and is further enhanced through RL training. Experimental results on the AIME and MATH benchmarks show that M1 not only outperforms previous linear RNN models but also matches the performance of state-of-the-art Deepseek R1 distilled reasoning models at a similar scale. We also compare our generation speed with a highly performant general purpose inference engine, vLLM, and observe more than a 3x speedup compared to a same size transformer. With throughput speedup, we are able to achieve higher accuracy compared to DeepSeek R1 distilled transformer reasoning models under a fixed generation time budget using self-consistency voting. Overall, we introduce a hybrid Mamba reasoning model and provide a more effective approach to scaling test-time generation using self-consistency or long chain of thought reasoning.

cs.LG↗

arXiv2025

OverFill: Two-Stage Models for Efficient Language Model Decoding

Large language models (LLMs) excel across diverse tasks but face significant deployment challenges due to high inference costs. LLM inference comprises prefill (compute-bound) and decode (memory-bound) stages, with decode dominating latency particularly for long sequences. Current decoder-only models handle both stages uniformly, despite their distinct computational profiles. We propose OverFill, which decouples these stages to optimize accuracy-efficiency tradeoffs. OverFill begins with a full model for prefill, processing system and user inputs in parallel. It then switches to a dense pruned model, while generating tokens sequentially. Leveraging more compute during prefill, OverFill improves generation quality with minimal latency overhead. Our 3B-to-1B OverFill configuration outperforms 1B pruned models by 83.2%, while the 8B-to-3B configuration improves over 3B pruned models by 79.2% on average across standard benchmarks. OverFill matches the performance of same-sized models trained from scratch, while using significantly less training data. Our code is available at https://github.com/friendshipkim/overfill.

cs.AI↗

arXiv2025

The Mamba in the Llama: Distilling and Accelerating Hybrid Models

Linear RNN architectures, like Mamba, can be competitive with Transformer models in language modeling while having advantageous deployment characteristics. Given the focus on training large-scale Transformer models, we consider the challenge of converting these pretrained models for deployment. We demonstrate that it is feasible to distill large Transformers into linear RNNs by reusing the linear projection weights from attention layers with academic GPU resources. The resulting hybrid model, which incorporates a quarter of the attention layers, achieves performance comparable to the original Transformer in chat benchmarks and outperforms open-source hybrid Mamba models trained from scratch with trillions of tokens in both chat benchmarks and general benchmarks. Moreover, we introduce a hardware-aware speculative decoding algorithm that accelerates the inference speed of Mamba and hybrid models. Overall we show how, with limited computation resources, we can remove many of the original attention layers and generate from the resulting model more efficiently. Our top-performing model, distilled from Llama3-8B-Instruct, achieves a 29.61 length-controlled win rate on AlpacaEval 2 against GPT-4 and 7.35 on MT-Bench, surpassing the best 8B scale instruction-tuned linear RNN model. We also find that the distilled model has natural length extrapolation, showing almost perfect accuracy in the needle-in-a-haystack test at 20x the distillation length. Code and pre-trained checkpoints are open-sourced at https://github.com/jxiw/MambaInLlama and https://github.com/itsdaniele/speculative_mamba.

cs.LG↗

arXiv2025

Thinking Slow, Fast: Scaling Inference Compute with Distilled Reasoners

Recent advancements have demonstrated that the performance of large language models (LLMs) can be significantly enhanced by scaling computational resources at test time. A common strategy involves generating multiple Chain-of-Thought (CoT) trajectories and aggregating their outputs through various selection mechanisms. This raises a fundamental question: can models with lower complexity leverage their superior generation throughput to outperform similarly sized Transformers for a fixed computational budget? To address this question and overcome the lack of strong subquadratic reasoners, we distill pure and hybrid Mamba models from pretrained Transformers. Trained on only 8 billion tokens, our distilled models show strong performance and scaling on mathematical reasoning datasets while being much faster at inference for large batches and long sequences. Despite the zero-shot performance hit due to distillation, both pure and hybrid Mamba models can scale their coverage and accuracy performance past their Transformer teacher models under fixed time budgets, opening a new direction for scaling inference compute.

cs.CL↗

arXiv2024

MambaByte: Token-free Selective State Space Model

Token-free language models learn directly from raw bytes and remove the inductive bias of subword tokenization. Operating on bytes, however, results in significantly longer sequences. In this setting, standard autoregressive Transformers scale poorly as the effective memory required grows with sequence length. The recent development of the Mamba state space model (SSM) offers an appealing alternative approach with a fixed-sized memory state and efficient decoding. We propose MambaByte, a token-free adaptation of the Mamba SSM trained autoregressively on byte sequences. In terms of modeling, we show MambaByte to be competitive with, and even to outperform, state-of-the-art subword Transformers on language modeling tasks while maintaining the benefits of token-free language models, such as robustness to noise. In terms of efficiency, we develop an adaptation of speculative decoding with tokenized drafting and byte-level verification. This results in a $2.6\times$ inference speedup to the standard MambaByte implementation, showing similar decoding efficiency as the subword Mamba. These findings establish the viability of SSMs in enabling token-free language modeling.

cs.CL↗