Large Language Models (LLMs) have demonstrated remarkable abilities in tackling a wide range of complex tasks. However, their huge computational and memory costs raise significant challenges in deploying these models on resource-constrained devices or efficiently serving them. Prior approaches have attempted to alleviate these problems by permanently removing less important model structures, yet these methods often result in substantial performance degradation due to the permanent deletion of model parameters. In this work, we tried to mitigate this issue by reducing the number of active parameters without permanently removing them. Specifically, we introduce a differentiable dynamic pruning method that pushes dense models to maintain a fixed number of active parameters by converting their MLP layers into a Mixture of Experts (MoE) architecture. Our method, even without fine-tuning, consistently outperforms previous structural pruning techniques across diverse model families, including Phi-2, LLaMA-2, LLaMA-3, and Qwen-2.5.

Large language models (LLMs) concentrate substantial knowledge in specialized domains due to extensive pretraining and instruction tuning, and they are now central to commercial and scientific practice. Yet access is usually limited to costly, rate-limited interfaces, which motivates methods that can extract targeted domain knowledge with minimal querying effort. A further challenge is that the target domain may be unknown in advance, so naive or generic prompts waste queries and fail to expose the underlying concepts and relations that structure the domain. In this work, we introduce a query-efficient approach for domain-specific knowledge stealing from black-box language models. Rather than issuing random questions or generic templates, our framework performs self-directed exploration that lets the model find the direction and mine domain knowledge by itself. Starting from a small and diverse seed, it discovers salient domain entities and induces their relations through structured question families that elicit definitional, functional, and compositional information. A feedback-driven controller analyzes the errors and uncertainty of the extracted surrogate model and uses this signal to refine subsequent queries, all without relying on prior domain knowledge or external resources. We evaluate the method in two expert-centric settings, medicine and finance, and observe consistently better performance while requiring significantly fewer queries.

Discrete motion tokenization has recently enabled Large Language Models (LLMs) to serve as versatile backbones for motion understanding and motion-language reasoning. However, existing pipelines typically decouple motion quantization from semantic embedding learning, linking them solely via token IDs. This approach fails to effectively align the intrinsic geometry of the motion space with the embedding space, thereby hindering the LLM's capacity for nuanced motion reasoning. We argue that alignment is most effective when both modalities share a unified geometric basis. Therefore, instead of forcing the LLM to reconstruct the complex geometry among motion tokens from scratch, we present a novel framework that explicitly enforces orthogonality on both the motion codebook and the LLM embedding space, ensuring that their relational structures naturally mirror each other. Specifically, we employ a decoder-only quantizer with Gumbel-Softmax for differentiable training and balanced codebook usage. To bridge the modalities, we use a sparse projection that maps motion codes into the LLM embedding space while preserving orthogonality. Finally, a two-stage orthonormal regularization schedule enforces soft constraints during tokenizer training and LLM fine-tuning to maintain geometric alignment without hindering semantic adaptation. Extensive experiments on HumanML3D demonstrate that our framework achieves a 20

Instruction-based text editing is increasingly critical for real-world applications such as code editors (e.g., Cursor), but Large Language Models (LLMs) continue to struggle with this task. Unlike free-form generation, editing requires faithfully implementing user instructions while preserving unchanged content, as even minor unintended modifications can break functionality. Existing approaches treat editing as generic text generation, leading to two key failures: they struggle to faithfully align edits with diverse user intents, and they often over-edit unchanged regions. We propose HyperEdit to address both issues. First, we introduce hypernetwork-based dynamic adaptation that generates request-specific parameters, enabling the model to tailor its editing strategy to each instruction. Second, we develop difference-aware regularization that focuses supervision on modified spans, preventing over-editing while ensuring precise, minimal changes. HyperEdit achieves a 9

Large Language Models (LLMs) demonstrate remarkable capabilities but face deployment challenges due to their high computational demands. Traditional pruning methods reduce these costs by permanently removing parameters, which inevitably leads to performance degradation. To mitigate this issue, we propose ToMoE, a method that transforms dense LLMs into Mixture-of-Experts (MoE) models by uncovering experts inherently present within dense models, without requiring any weight updates. ToMoE leverages dynamic structural pruning to unify expert construction and router training in a single stage, achieving consistently strong performance. Remarkably, even without fine-tuning \revise{the model weights}, ToMoE consistently outperforms state-of-the-art pruning and MoE techniques across Phi-2, LLaMA-2, LLaMA-3, and Qwen-2.5 models. The code for this paper is available at https://github.com/gaosh/ToMoE.

Parameter-efficient fine-tuning (PEFT) enables efficient adaptation of large language models, but existing MoE-based PEFT methods typically improve capacity by storing multiple full LoRA experts, causing adapter storage to grow linearly with the number of experts and restricting adaptation to a fixed expert pool. We ask whether MoE-based PEFT can produce instance-specific adaptations without explicitly storing a separate LoRA module for each expert. To address this gap, we propose MoEGen, an adaptation framework that shifts MoE-based PEFT from expert selection to expert-conditioned parameter generation. Instead of storing each expert as a full LoRA adapter, MoEGen represents each expert as a small learnable vector, termed an expert code. It routes each input over these vectors and uses their weighted combination to condition a lightweight hypernetwork that generates input-specific low-rank updates. This design decouples expert capacity from adapter storage while enabling instance-conditioned adaptation. Experiments on eight commonsense reasoning benchmarks show consistent improvements over strong static and MoE-based PEFT baselines across three backbones. MoEGen also performs strongly in joint medical and legal-domain adaptation.

The ability to recognize one's own limitations and decide whether to solve a problem or delegate is fundamental for reliable intelligent systems. Yet we show that modern large language models systematically lack this ability: across diverse model families and scales, they overestimate their competence and attempt queries they cannot solve. We refer to this ability as Capability Self-Assessment (CSA) and formulate it as a policy-learning problem, aiming to improve self-assessment while preserving the model's original capabilities. Our results show that reinforcement learning teaches CSA effectively, significantly outperforming supervised fine-tuning while preserving original capabilities. In contrast, supervised fine-tuning severely degrades the capabilities the model is meant to assess. Moreover, learned self-assessment behavior generalizes well out of distribution, suggesting that CSA is a transferable model trait. Finally, CSA is practically useful: it improves local-cloud decision making at inference time and provides a signal for targeted data selection during training.

Group Relative Policy Optimization (GRPO) has become the dominant method for reinforcement learning with verifiable rewards in large language models, but it suffers from two critical limitations: gradient vanishing and diversity collapse. When training questions are too easy or too hard, all sampled responses receive identical rewards, yielding zero gradients. Meanwhile, the model tends to collapse its responses toward a single reasoning pattern rather than exploring diverse strategies. We propose Transformation-Augmented GRPO (TA-GRPO), a simple but effective method that addresses both issues via question rephrasing. For each training question, we automatically generate multiple problem-equivalent rephrasings that alter wording, format, and information order while preserving the underlying meaning. Because these rephrasings shift the model's perceived difficulty, pooling responses across the original and its rephrasings yields mixed rewards and more diverse reasoning paths. TA-GRPO jointly computes advantages over this expanded response set and aligns all importance ratios to the original question, enabling the model to learn from a richer set of solution attempts. Experiments on four LLMs (Qwen3-1.7B, Qwen3-4B, Llama-3.2-1B, Llama-3.2-3B) show that TA-GRPO consistently improves pass@k on competition-level benchmarks (AMC, OlympiadBench, AIME24, AIME25) and out-of-distribution benchmarks (Minerva, GPQA-Diamond). Notably, it improves the average pass@32 of Qwen3-1.7B and Qwen3-4B by 4.97 and 4.34 points, respectively, and matches the exploration quality of baselines trained on up to 2.5× more data.

Grounding human motion in language, and language in motion, is a central step toward physical AI systems that can understand, generate, and communicate human behavior. Unified motion–language systems first coupled the two directions through a shared discrete motion codebook, but quantization limits generation quality. The strongest generators buy quality back at growing cost: stacked residual codebooks enlarge the representation; masked decoding stages, long autoregressive rollouts, and denoising chains of tens to hundreds of steps stretch inference; even the continuous-latent designs among them reach their latent only through an iterative diffusion head; and none of this decoding machinery serves understanding. We therefore propose MUGEN, a unified motion–language framework that pays neither cost: no codebook, one draw. A single adaptive-length autoencoder compresses any-length motion into a few continuous latent slots, the system's only motion representation: the language model generates them for text-to-motion and reads them back for motion understanding. Depth-routed hidden states let each slot read from the transformer depth it needs, and a calibrated head predicts a joint distribution over the full latent set, so a single draw carries the text-conditional, cross-slot variation a description permits. At a decoding cost of K language-model steps, one draw, and one decoder pass, MUGEN leads language-model baselines on FID on HumanML3D while raising retrieval precision above the real-motion reference under the standard evaluator, achieves the best CIDEr and BLEU@4 scores, and surpasses the discrete-token state of the art on every retrieval and alignment metric on SnapMoGen.

Large language model (LLM) routing has emerged as a critical strategy to balance model performance and cost-efficiency by dynamically selecting services from various model providers. However, LLM routing adds an intermediate layer between users and LLMs, creating new privacy risks to user data. These privacy risks have not been systematically studied. Although cryptographic techniques such as Secure Multi-Party Computation (MPC) enable privacy-preserving computation, their protocol design and implementation remain under-explored, and naïve implementations typically incur prohibitive computational overhead. To address this, we propose a privacy-preserving LLM routing framework (PPRoute). PPRoute includes multiple strategies to speed up encoder inference and nearest neighbor search under the MPC and maintain the quality of LLM routing. First, PPRoute uses MPC-friendly operations to boost the encoder inference. Second, PPRoute uses a multiple-step model training algorithm to maintain routing quality despite the constraints of the encrypted domain. Third, PPRoute proposes an unsorted Top-k algorithm with O(1) communication complexity for secure sorting in model search, significantly reducing communication latency. Across different datasets, PPRoute achieves the performance of plaintext counterparts, while achieving approximately a 20× speedup over naïve MPC implementations.

Federated Learning (FL) enables collaborative model training across institutions without sharing raw data. However, gradient sharing still risks privacy leakage, such as gradient inversion attacks. Homomorphic Encryption (HE) can secure aggregation but often incurs prohibitive computational and communication overhead. Existing HE-based FL methods sit at two extremes: encrypting all gradients for full privacy at high cost, or partially encrypting gradients to save resources while exposing vulnerabilities. We present **DictPFL**, a practical framework that achieves full gradient protection with minimal overhead. DictPFL encrypts every transmitted gradient while keeping non-transmitted parameters local, preserving privacy without heavy computation. It introduces two key modules: **Decompose-for-Partial-Encrypt (DePE)**, which decomposes model weights into a static dictionary and an updatable lookup table—only the latter is encrypted and aggregated, while the static dictionary remains local and requires neither sharing nor encryption; and **Prune-for-Minimum-Encrypt (PrME)**, which applies encryption-aware pruning to minimize encrypted parameters via consistent, history-guided masks. Experiments show that DictPFL reduces communication cost by 402-748$\times$ and accelerates training by 28-65$\times$ compared to fully encrypted FL, while outperforming state-of-the-art selective encryption methods by 51-155$\times$ in overhead and 4-19$\times$ in speed. Remarkably, DictPFL’s runtime is within 2$\times$ of plaintext FL, demonstrating, for the first time, that HE-based private federated learning is practical for real-world deployment. The code is publicly available at https://github.com/UCF-ML-Research/DictPFL.

Large Language Models (LLMs) have significantly advanced AI with their exceptional performance across a wide range of tasks. However, their extensive computational requirements restrict their use on devices with limited resources.While recent compression methods based on low-rank matrices show potentialsolutions, they often suffer from significant loss of accuracy or introduce substantialoverhead in parameters and inference time. In this paper, we introduce Modular De-composition (MoDeGPT), a new, efficient, and structured compression frameworkthat overcomes these limitations. MoDeGPT jointly decomposes pairs of consecu-tive subcomponents within Transformer blocks, reduces hidden dimensions throughoutput reconstruction on a larger structural scale than conventional low-rank meth-ods, and repurposes three classical matrix decomposition algorithms—Nyströmapproximation, CR decomposition, and SVD—to ensure bounded errors in ournovel decomposition approach. Our experiments show that MoDeGPT, withoutrelying on backward propagation, consistently matches or surpasses the performance of prior techniques that depend on gradient information, while achieving a98% reduction in compute costs when compressing a 13B-parameter model. OnLLaMA-2/3 and OPT models, MoDeGPT retains 90-95% of zero-shot performancewith compression rates of 25-30%. The compression process can be completed ona single GPU in a few hours, boosting inference throughput by up to 46%.

The evolution of pre-trained large language models (LLMs) has significantly transformed natural language processing. However, these advancements pose challenges, particularly the unintended memorization of training data, which raises ethical and privacy concerns. While prior research has largely focused on mitigating memorization or extracting memorized information, the deliberate control of memorization has been underexplored. This study addresses this gap by introducing a novel and unified gradient-based weight pruning framework to freely control memorization rates in LLMs. Our method enables fine-grained control over pruning parameters, allowing models to suppress or enhance memorization based on application-specific requirements. Experimental results demonstrate that our approach effectively balances the trade-offs between memorization and generalization, with an increase of up to 89.3% in Fractional ER suppression and 40.9% in Exact ER amplification compared to the original models.

Code generation is increasingly critical for real-world applications. Still, diffusion-based large language models continue to struggle with this demand. Unlike free-form text, code requires syntactic precision; even minor structural inconsistencies can render a program non-executable. Existing diffusion-based large language models rely on random token masking for corruption, leading to two key failures: they lack awareness of syntactic boundaries during the iterative denoising process, and they fail to capture the long-range hierarchical dependencies essential for program correctness. We propose TreeDiff to address both issues. Specifically, we propose a syntax-aware diffusion framework that incorporates structural priors from Abstract Syntax Tree (AST) into the corruption process. Instead of masking individual tokens at random, we selectively mask tokens belonging to key AST nodes. By aligning the corruption process with the underlying structure of code, our method encourages the model to internalize the compositional nature of programming languages, enabling it to reconstruct programs that respect grammatical boundaries and capture long-range dependencies. Our method achieves a 13.3

The rapid proliferation of large language models (LLMs) in natural language processing (NLP) has created a critical need for techniques that enable efficient deployment on memory-constrained devices without compromising performance. We present a method to prune LLMs that selectively prunes model blocks based on an importance score and replaces them with a low-parameter replacement strategy. Specifically, we propose a principled metric to replace each pruned block using a weight-sharing mechanism that leverages unpruned counterparts from the model and block-specific low-rank adapters. Furthermore, we facilitate the learning of these replacement blocks with output feature normalization and an adapter initialization scheme built on low-rank SVD reconstructions. Empirical evaluations demonstrate substantial performance gains over existing methods, achieving state-of-the-art performance on 5/6 benchmarks for a compression rate of 30% and 6/6 benchmarks for a compression rate of 40%. We also demonstrate that our approach can extend smaller models, boosting performance on 6/6 benchmarks using only approximate to 0.3% tokens of extended training with minimal additional parameter costs.

LLM-as-a-Judge has emerged as a promising tool for automatically evaluating generated outputs, but its reliability is often undermined by potential biases in judgment. Existing efforts to mitigate these biases face key limitations: in-context learning-based methods fail to address rooted biases due to the evaluator’s limited capacity for self-reflection, whereas fine-tuning is not applicable to all evaluator types, especially closed-source models. To address this challenge, we introduce the **R**easoning-based **B**ias **D**etector (RBD), which is a plug-in module that identifies biased evaluations and generates structured reasoning to guide evaluator self-correction. Rather than modifying the evaluator itself, RBD operates externally and engages in an iterative process of bias detection and feedback-driven revision. To support its development, we design a complete pipeline consisting of biased dataset construction, supervision collection, distilled reasoning-based fine-tuning of RBD, and integration with LLM evaluators. We fine-tune four sizes of RBD models, ranging from 1.5B to 14B, and observe consistent performance improvements across all scales. Experimental results on 4 bias types—verbosity, position, bandwagon, and sentiment—evaluated using 8 LLM evaluators demonstrate RBD’s strong effectiveness. For example, the RBD-8B model improves evaluation accuracy by an average of 18.5% and consistency by 10.9%, and surpasses prompting-based baselines and fine-tuned judges by 12.8% and 17.2%, respectively. These results highlight RBD’s effectiveness and scalability. Additional experiments further demonstrate its strong generalization across biases and domains, as well as its efficiency.

Text-to-image (T2I) diffusion models have demonstrated impressive image generation capabilities. Still, their computational intensity prohibits resource-constrained organizations from deploying T2I models after fine-tuning them on their internal *target* data. While pruning techniques offer a potential solution to reduce the computational burden of T2I models, static pruning methods use the same pruned model for all input prompts, overlooking the varying capacity requirements of different prompts. Dynamic pruning addresses this issue by utilizing a separate sub-network for each prompt, but it prevents batch parallelism on GPUs. To overcome these limitations, we introduce Adaptive Prompt-Tailored Pruning (APTP), a novel prompt-based pruning method designed for T2I diffusion models. Central to our approach is a *prompt router* model, which learns to determine the required capacity for an input text prompt and routes it to an architecture code, given a total desired compute budget for prompts. Each architecture code represents a specialized model tailored to the prompts assigned to it, and the number of codes is a hyperparameter. We train the prompt router and architecture codes using contrastive learning, ensuring that similar prompts are mapped to nearby codes. Further, we employ optimal transport to prevent the codes from collapsing into a single one. We demonstrate APTP's effectiveness by pruning Stable Diffusion (SD) V2.1 using CC3M and COCO as *target* datasets. APTP outperforms the single-model pruning baselines in terms of FID, CLIP, and CMMD scores. Our analysis of the clusters learned by APTP reveals they are semantically meaningful. We also show that APTP can automatically discover previously empirically found challenging prompts for SD, *e.g.,* prompts for generating text images, assigning them to higher capacity codes.

We study cost-aware routing for large language models across diverse and dynamic pools of models. Existing approaches often overlook prompt-specific context, rely on expensive model profiling, assume a fixed set of experts, or use inefficient trial-and-error strategies. We introduce Cost-Spectrum Contrastive Routing (CSCR), a lightweight framework that maps both prompts and models into a shared embedding space to enable fast, cost-sensitive selection. CSCR uses compact, fast-to-compute logit footprints for open-source models and perplexity fingerprints for black-box APIs. A contrastive encoder is trained to favor the cheapest accurate expert within adaptive cost bands. At inference time, routing reduces to a single $k$‑NN lookup via a FAISS index, requiring no retraining when the expert pool changes and enabling microsecond latency. Across multiple benchmarks, CSCR consistently outperforms baselines, improving the accuracy–cost tradeoff by up to 25\%, while generalizing robustly to unseen LLMs and out-of-distribution prompts.

Although LLMs have achieved significant success, their reliance on large volumes of human-annotated data has limited their potential for further scaling. In this situation, utilizing self-generated synthetic data has become crucial for fine-tuning LLMs without extensive human annotation. However, current methods often fail to ensure consistent improvements across iterations, with performance stagnating after only minimal updates. To overcome these challenges, we introduce Dynamic Noise Preference Optimization (DNPO), which combines dynamic sample labeling for constructing preference pairs with controlled, trainable noise injection during preference optimization. Our approach effectively prevents stagnation and enables continuous improvement. In experiments with Llama-3.2-3B and Zephyr-7B, DNPO consistently outperforms existing methods across multiple benchmarks. Additionally, with Zephyr-7B, DNPO shows a significant improvement in model-generated data quality, with a 29.4

Recent advances in diffusion generative models have yielded remarkable progress. While the quality of generated content continues to improve, these models have grown considerably in size and complexity. This increasing computational burden poses significant challenges, particularly in resource-constrained deployment scenarios such as mobile devices. The combination of model pruning and knowledge distillation has emerged as a promising solution to reduce computational demands while preserving generation quality. However, this technique inadvertently propagates undesirable behaviors, including the generation of copyrighted content and unsafe concepts, even when such instances are absent from the fine-tuning dataset. In this paper, we propose a novel bilevel optimization framework for pruned diffusion models that consolidates the fine-tuning and unlearning processes into a unified phase. Our approach maintains the principal advantages of distillation-namely, efficient convergence and style transfer capabilities-while selectively suppressing the generation of unwanted content. This plug-in framework is compatible with various pruning and concept unlearning methods, facilitating efficient, safe deployment of diffusion models in controlled environments. Code is available here.