While model-based verifiers are essential for scaling Reinforcement Learning with Verifiable Rewards (RLVR), current outcome-centric verification paradigms primarily focus on the consistency between the final result and the ground truth, often neglecting potential errors in the derivation process. This leads to assigning positive rewards to correct answers produced from incorrect derivations. To bridge this gap, we introduce **PRIME**, a benchmark for evaluating verifiers on **PR**ocess-outcome alignment verification **I**n **M**athematics and **E**ngineering. Curated from a comprehensive collection of college-level STEM problems, **PRIME** comprises 2,530 high-difficulty samples through a consistency-based filtering pipeline. Through extensive evaluation, we find that current verifiers frequently fail to detect derivation flaws. Furthermore, we propose a process-aware RLVR training paradigm utilizing verifiers selected via **PRIME**. This approach substantially outperforms the outcome-only verification baseline, achieving absolute performance gains of **8.29%**, **9.12%**, and **7.31%** on AIME24, AIME25, and Beyond-AIME, respectively, for the Qwen3-14B-Base model. Finally, we demonstrate a strong linear correlation (R² > 0.92) between verifier accuracy on **PRIME** and RLVR training effectiveness, validating **PRIME** as a reliable predictor for verifier selection.

Autoregressive (AR) decoding in large language models (LLMs) is latency-bounded by strictly sequential token generation.Speculative decoding mitigates this bottleneck by letting a fast drafter propose multi-token candidates that are then verified in parallel by the target model; yet most existing systems still rely on AR drafters, limiting wall-clock gains.We present **DiffuSpec**, which repurposes a *diffusion language model* (DLM) as a *parallel* drafter to generate multi-token proposals in a single forward pass while remaining compatible with standard AR verifiers.However, DLM drafting presents unique challenges: 1) bidirectional conditioning produces a token lattice where locally optimal tokens may fail to form a valid causal sequence; 2) the mechanism requires tuning the draft length, which induces a speed–quality trade-off. To address these issues, we introduce (i) *Causal-consistency Path Search* (CPS) to extract verifier-aligned causal paths from the lattice, and (ii) an *Adaptive Draft-Length* (ADL) controller that adjusts proposal lengths using online acceptance feedback.Across benchmarks, DiffuSpec achieves up to 3× wall-clock speedup and consistently outperforms strong baselines, demonstrating diffusion-based drafting as a competitive alternative to AR drafters for speculative decoding.

Resource constraints often limit the parameter capacity of Large Language Models (LLMs), thereby hindering their performance. Although existing approaches leverage parameter sharing to reuse a fixed set of parameters within constrained budgets, they typically require each layer to fulfill multiple roles over a fixed number of iterations. This design compromises both efficiency and adaptability. In this work, we propose the **Zero Token Transformer (ZTT)**, which employs a head-tail decoupled parameter cycling strategy. Specifically, we decouple the first (head) and last (tail) layers from the parameter cycling process, enabling iterative refinement solely within the intermediate layers. Furthermore, we introduce a Zero-Token Mechanism, wherein a virtual token with a trainable key and a zero-valued vector functions as a standard token. The resulting attention scores not only reflect the computational significance of each layer but also facilitate dynamic early exiting, thereby preserving overall model accuracy. Our approach achieves superior performance under strict parameter constraints, substantially reduces computational overhead via early exits, and can be seamlessly integrated into the fine-tuning of existing pre-trained models, improving both efficiency and adaptability.

Large language models (LLMs) excel in natural language processing but demand intensive computation. To mitigate this, various quantization methods have been explored, yet they compromise LLM performance. This paper unveils a previously overlooked type of outliers in LLMs. Such outliers are found to allocate most of the attention scores on initial tokens of input, termed as pivot tokens, which are crucial to the performance of quantized LLMs. Given that, we propose IntactKV to generate the KV cache of pivot tokens losslessly from the full-precision model. The approach is simple and easy to combine with existing quantization solutions with no extra inference overhead. Besides, IntactKV can be calibrated as additional LLM parameters to boost the quantized LLMs further with minimal training costs. Mathematical analysis also proves that IntactKV effectively reduces the upper bound of quantization error. Empirical results show that IntactKV brings consistent improvement over various quantization methods across different LLMs and downstream tasks, leading to the new state-of-the-art for LLM quantization. The codes are available at https://github.com/ruikangliu/IntactKV.

Grammatical error correction (GEC) aims to correct errors in given sentences and is significant to many downstream natural language understanding tasks. Recent work introduces the idea of grammatical error detection (GED) to improve the GEC task performance. In contrast, these explicit multi-stage works propagate and amplify the problem of misclassification of the GED module. To introduce more convincing error type information, we propose an end-to-end framework in this paper, which Leverages Error Type (LET) information in the generation process. First, the input text is fed into a classification module to obtain the error type corresponding to each token. Then, we introduce the category information into the decoder’s input and cross-attention module in two ways, respectively. Experiments on various datasets show that our proposed method outperforms existing methods by a clear margin.

It has been commonly observed that a teacher model with superior performance does not necessarily result in a stronger student, highlighting a discrepancy between current teacher training practices and effective knowledge transfer. In order to enhance the guidance of the teacher training process, we introduce the concept of distillation influence to determine the impact of distillation from each training sample on the student’s generalization ability. In this paper, we propose Learning Good Teacher Matters (LGTM), an efficient training technique for incorporating distillation influence into the teacher’s learning process. By prioritizing samples that are likely to enhance the student’s generalization ability, our LGTM outperforms 10 common knowledge distillation baselines on 6 text classification tasks in the GLUE benchmark.

Syntactic structure has long been argued to be potentially useful for enforcing accurate word alignment and improving generalization performance of machine translation. Unfortunately, existing wisdom demonstrates its significance by considering only the syntactic structure of source tokens, neglecting the rich structural information from target tokens and the structural similarity between the source and target sentences. In this work, we propose to incorporate the syntactic structure of both source and target tokens into the encoder-decoder framework, tightly correlating the internal logic of word alignment and machine translation for multi-task learning. Particularly, we won’t leverage any annotated syntactic graph of the target side during training, so we introduce Dynamic Graph Convolution Networks (DGCN) on observed target tokens to sequentially and simultaneously generate the target tokens and the corresponding syntactic graphs, and further guide the word alignment. On this basis, Hierarchical Graph Random Walks (HGRW) are performed on the syntactic graphs of both source and target sides, for incorporating structured constraints on machine translation outputs. Experiments on four publicly available language pairs verify that our method is highly effective in capturing syntactic structure in different languages, consistently outperforming baselines in alignment accuracy and demonstrating promising results in translation quality.

Recent image captioning models have made much progress for exploring the multi-modal interaction, such as attention mechanisms. Though these mechanisms can boost the interaction, there are still two gaps between the visual and language domains: (1) the gap between the visual features and textual semantics, (2) the gap between the disordering of visual features and the ordering of texts. To bridge the gaps we propose a high-level semantic planning (HSP) mechanism that incorporates both a semantic reconstruction and an explicit order planning. We integrate the planning mechanism to the attention based caption model and propose the High-level Semantic PLanning based Attention Network (HS-PLAN). First, an attention based reconstruction module is designed to reconstruct the visual features with high-level semantic information. Then we apply a pointer network to serialize the features and obtain the explicit order plan to guide the generation. Experiments conducted on MS COCO show that our model outperforms previous methods and achieves the state-of-the-art performance of 133.4% CIDEr-D score.