ICML 2025 Papers

Post-training quantization (PTQ) is a promising solution for deploying large language models (LLMs) on resource-constrained devices. Early methods developed for small-scale networks, such as ResNet, rely on gradient-based optimization, which becomes impractical for hyper-scale LLMs with billions of parameters. While recently proposed backpropagation-free or transformation-based methods alleviate this issue, they ignore inter-layer interactions or use the naive nearest-rounding-based quantized weight assignment to save the heavy computational cost of weight optimization. In this paper, we introduce a novel backpropagation-free PTQ algorithm that optimizes quantized weights by considering inter-layer dependencies. The key innovation is the development of attention-aware Hessian matrices that capture inter-layer interactions within the attention module. Extensive experiments demonstrate that our approach not only outperforms existing weight quantization methods but also shows good synergy with conventional methods to suppress activation outliers, leading to state-of-the-art weight-activation quantization performance. The code will be available at https://github.com/SamsungLabs/BoA.

Foundation models trained on patient electronic health records (EHRs) require tokenizing medical data into sequences of discrete vocabulary items. Existing tokenizers treat medical codes from EHRs as isolated textual tokens. However, each medical code is defined by its textual description, its position in ontological hierarchies, and its relationships to other codes, such as disease co-occurrences and drug-treatment associations. Medical vocabularies contain more than 600,000 codes with critical information for clinical reasoning. We introduce MedTok, a multimodal medical code tokenizer that uses the text descriptions and relational context of codes. MedTok processes text using a language model encoder and encodes the relational structure with a graph encoder. It then quantizes both modalities into a unified token space, preserving modality-specific and cross-modality information. We integrate MedTok into five EHR models and evaluate it on operational and clinical tasks across in-patient and out-patient datasets, including outcome prediction, diagnosis classification, drug recommendation, and risk stratification. Swapping standard EHR tokenizers with MedTok improves AUPRC across all EHR models, by 4.10% on MIMIC-III, 4.78% on MIMIC-IV, and 11.32% on EHRShot, with the largest gains in drug recommendation. Beyond EHR modeling, we demonstrate using MedTok tokenizer with medical QA systems. Our results demonstrate the potential of MedTok as a unified tokenizer for medical codes, improving tokenization for medical foundation models.

Kronecker-sparse (KS) matrices—whose supports are Kronecker products of identity and all-ones blocks—underpin the structure of Butterfly and Monarch matrices and offer the promise of more efficient models. However, existing GPU kernels for KS matrix multiplication suffer from high data movement costs, with up to 50% of time spent on memory-bound tensor permutations. We propose a fused, output-stationary GPU kernel that eliminates these overheads, reducing global memory traffic threefold. Across 600 KS patterns, our kernel achieves in FP32 a median speedup of x1.4 and lowers energy consumption by 15%. A simple heuristic based on KS pattern parameters predicts when our method outperforms existing ones. We release all code at [github.com/PascalCarrivain/ksmm](https://github.com/PascalCarrivain/ksmm), including a PyTorch-compatible *KSLinear* layer, and demonstrate in FP32 end-to-end latency reductions of up to 22% in ViT-S/16 and 16% in GPT-2 medium.

AI for software engineering has made remarkable progress, becoming a notable success within generative AI. Despite this, achieving fully automated software engineering is still a significant challenge, requiring research efforts across both academia and industry. In this position paper, our goal is threefold. First, we provide a taxonomy of measures and tasks to categorize work towards AI software engineering. Second, we outline the key bottlenecks permeating today's approaches. Finally, we highlight promising paths towards making progress on these bottlenecks to guide future research in this rapidly maturing field.

Numerous studies have assessed the proficiency of AI systems, particularly large language models (LLMs), in facilitating everyday tasks such as email writing, question answering, and creative content generation. However, researchers face unique challenges and opportunities in leveraging LLMs for their own work, such as brainstorming research ideas, designing experiments, and writing or reviewing papers. In this study, we introduce AAAR-1.0, a benchmark dataset designed to evaluate LLM performance in three fundamental, expertise-intensive research tasks: (i) EquationInference, assessing the correctness of equations based on the contextual information in paper submissions; (ii) ExperimentDesign, designing experiments to validate research ideas and solutions; and (iii) PaperWeakness, identifying weaknesses in paper submissions. AAAR-1.0 differs from prior benchmarks in two key ways: first, it is explicitly research-oriented, with tasks requiring deep domain expertise; second, it is researcher-oriented, mirroring the primary activities that researchers engage in on a daily basis. An evaluation of both open-source and proprietary LLMs reveals their potential as well as limitations in conducting sophisticated research tasks. We will release the AAAR-1.0 and keep iterating it to new versions.

Existing methods for code generation use code snippets as seed data, restricting the complexity and diversity of the synthesized data. In this paper, we introduce a novel feature tree-based synthesis framework, which revolves around hierarchical code features derived from high-level abstractions of code. The feature tree is constructed from raw data and refined iteratively to increase the quantity and diversity of the extracted features, which captures and recognizes more complex patterns and relationships within the code. By adjusting the depth and breadth of the sampled subtrees, our framework provides precise control over the complexity of the generated code, enabling functionalities that range from function-level operations to multi-file scenarios. We fine-tuned widely-used base models to obtain EpiCoder series, achieving state-of-the-art performance on multiple benchmarks at both the function and file levels. In particular, empirical evidence indicates that our approach shows significant potential in the synthesizing of repository-level code data. Our code and data are publicly available.

Transformer-based large language models (LLMs) have already achieved remarkable results on long-text tasks, but the limited GPU memory (VRAM) resources struggle to accommodate the linearly growing demand for key-value (KV) cache as the sequence length increases, which has become a bottleneck for the application of LLMs on long sequences. Existing KV cache compression methods include eviction, merging, or quantization of the KV cache to reduce its size. However, compression results in irreversible information forgetting, potentially affecting the accuracy of subsequent decoding. In this paper, we propose SpeCache, which takes full advantage of the large and easily expandable CPU memory to offload the complete KV cache, and dynamically fetches KV pairs back in each decoding step based on their importance measured by low-precision KV cache copy in VRAM. To avoid inference latency caused by CPU-GPU communication, SpeCache speculatively predicts the KV pairs that the next token might attend to, allowing us to prefetch them before the next decoding step which enables parallelization of prefetching and computation. Experiments on LongBench and Needle-in-a-Haystack benchmarks verify that SpeCache effectively reduces VRAM usage while avoiding information forgetting for long sequences without re-training, even with a 10x high KV cache compression ratio.

We introduce **SWE-Flow**, a novel data synthesis framework grounded in Test-Driven Development (TDD). Unlike existing software engineering data that rely on human-submitted issues, **SWE-Flow** automatically infers incremental development steps directly from unit tests, which inherently encapsulate high-level requirements. The core of **SWE-Flow** is the construction of a Runtime Dependency Graph (RDG), which precisely captures function interactions, enabling the generation of a structured, step-by-step *development schedule*. At each step, **SWE-Flow** produces a partial codebase, the corresponding unit tests, and the necessary code modifications, resulting in fully verifiable TDD tasks. With this approach, we generated 16,061 training instances and 2,020 test instances from real-world GitHub projects, creating the **SWE-Flow-Eval** benchmark. Our experiments show that fine-tuning open model on this dataset significantly improves performance in TDD-based coding. To facilitate further research, we release all code, datasets, models, and Docker images at [Github](https://github.com/Hambaobao/SWE-Flow).

Graph Edit Distance (GED) is a widely used metric for measuring similarity between two graphs. Computing the optimal GED is NP-hard, leading to the development of various neural and non-neural heuristics. While neural methods have achieved improved approximation quality compared to non-neural approaches, they face significant challenges: (1) They require large amounts of ground truth data, which is itself NP-hard to compute. (2) They operate as black boxes, offering limited interpretability. (3) They lack cross-domain generalization, necessitating expensive retraining for each new dataset. We address these limitations with GRAIL, introducing a paradigm shift in this domain. Instead of training a neural model to predict GED, GRAIL employs a novel combination of large language models (LLMs) and automated prompt tuning to generate a *program* that is used to compute GED. This shift from predicting GED to generating programs imparts various advantages, including end-to-end interpretability and an autonomous self-evolutionary learning mechanism without ground-truth supervision. Extensive experiments on seven datasets confirm that GRAIL not only surpasses state-of-the-art GED approximation methods in prediction quality but also achieves robust cross-domain generalization across diverse graph distributions.

Analog circuit topology synthesis is integral to Electronic Design Automation (EDA), enabling the automated creation of circuit structures tailored to specific design requirements. However, the vast design search space and strict constraint adherence make efficient synthesis challenging. Leveraging the versatility of Large Language Models (LLMs), we propose AUTOCIRCUIT-RL, a novel reinforcement learning (RL)-based framework for automated analog circuit synthesis. The framework operates in two phases: instruction tuning, where an LLM learns to generate circuit topologies from structured prompts encoding design constraints, and RL refinement, which further improves the instruction-tuned model using reward models that evaluate validity, efficiency, and output voltage. The refined model is then used directly to generate topologies that satisfy the design constraints. Empirical results show that AUTOCIRCUIT-RL generates ~12% more valid circuits and improves efficiency by ~14% compared to the best baselines, while reducing duplicate generation rates by ~38%. It achieves over 60% success in synthesizing valid circuits with limited training data, demonstrating strong generalization. These findings highlight the framework's effectiveness in scaling to complex circuits while maintaining efficiency and constraint adherence, marking a significant advancement in AI-driven circuit design.

Contemporary evaluation techniques are inadequate for agentic systems. These approaches either focus exclusively on final outcomes---ignoring the step-by-step nature of the thinking done by agentic systems---or require excessive manual labour. To address this, we introduce the **Agent-as-a-Judge** framework, wherein agentic systems are used to evaluate agentic systems. This is a natural extension of the LLM-as-a-Judge framework, incorporating agentic features that enable intermediate feedback for the entire task-solving processes for more precise evaluations. We apply the Agent-as-a-Judge framework to the task of code generation. To overcome issues with existing benchmarks and provide a proof-of-concept testbed for Agent-as-a-Judge, we present **DevAI**, a new benchmark of 55 realistic AI code generation tasks. DevAI includes rich manual annotations, like a total of 365 hierarchical solution requirements, which make it particularly suitable for an agentic evaluator. We benchmark three of the top code-generating agentic systems using Agent-as-a-Judge and find that our framework dramatically outperforms LLM-as-a-Judge and is as reliable as our human evaluation baseline. Altogether, we believe that this work represents a concrete step towards enabling vastly more sophisticated agentic systems. To help that, our dataset and the full implementation of Agent-as-a-Judge will be publically available at https://github.com/metauto-ai/agent-as-a-judge

The next-coordinate prediction paradigm has emerged as the de facto standard in current auto-regressive mesh generation methods. Despite their effectiveness, there is no efficient measurement for the various tokenizers that serialize meshes into sequences. In this paper, we introduce a new metric Per-Token-Mesh-Entropy (PTME) to evaluate the existing mesh tokenizers theoretically without any training. Building upon PTME, we propose a plug-and-play tokenization technique called coordinate merging. It further improves the compression ratios of existing tokenizers by rearranging and merging the most frequent patterns of coordinates. Through experiments on various tokenization methods like MeshXL, MeshAnything V2, and Edgerunner, we further validate the performance of our method. We hope that the proposed PTME and coordinate merging can enhance the existing mesh tokenizers and guide the further development of native mesh generation.

Inference latency stands as a critical bottleneck in the large-scale deployment of Large Language Models (LLMs). Speculative decoding methods have recently shown promise in accelerating inference without compromising the output distribution. However, existing work typically relies on a dualistic draft-verify framework and lacks rigorous theoretical grounding. In this paper, we introduce a novel \emph{polybasic} speculative decoding framework, underpinned by a comprehensive theoretical analysis. Specifically, we prove a fundamental theorem that characterizes the optimal inference time for multi-model speculative decoding systems, shedding light on how to extend beyond the dualistic approach to a more general polybasic paradigm. Through our theoretical investigation of multi-model token generation, we expose and optimize the interplay between model capabilities, acceptance lengths, and overall computational cost. Our framework supports both standalone implementation and integration with existing speculative techniques, leading to accelerated performance in practice. Experimental results across multiple model families demonstrate that our approach yields speedup ratios ranging from $3.31\times$ to $4.01\times$ for LLaMA2-Chat 7B, up to $3.87 \times$ for LLaMA3-8B, up to $4.43 \times$ for Vicuna-7B and up to $3.85 \times$ for Qwen2-7B---all while preserving the original output distribution. We release our theoretical proofs and implementation code to facilitate further investigation into polybasic speculative decoding.

One approach to reducing the massive costs of large language models (LLMs) is the use of quantized or sparse representations for training or deployment. While post-training compression methods are very popular, the question of obtaining even more accurate compressed models by *directly training* over such representations, i.e., *Quantization-Aware Training (QAT)*, is still open: for example, a recent study put the "optimal" bit-width at which models can be trained using QAT, while staying accuracy-competitive with standard FP16/BF16 precision, at 8-bits weights and activations. We advance this state-of-the-art via a new method called QuEST, for which we demonstrate optimality at 4-bits and stable convergence as low as 1-bit weights and activations. QuEST achieves this by improving two key aspects of QAT methods: (1) accurate and fast quantization of the (continuous) distributions of weights and activations via Hadamard normalization and MSE-optimal fitting; (2) a new *trust gradient estimator* based on the idea of explicitly minimizing the error between the noisy gradient computed over quantized states and the "true" (but unknown) full-precision gradient. Experiments on Llama-type architectures show that QuEST induces stable scaling laws across the entire range of hardware-supported precisions, and can be extended to sparse representations. We provide GPU kernel support showing that models produced by QuEST can be executed efficiently. Our code is available at [https://github.com/IST-DASLab/QuEST](https://github.com/IST-DASLab/QuEST).

Streaming generation models are utilized across fields, with the Transducer architecture being popular in industrial applications. However, its input-synchronous decoding mechanism presents challenges in tasks requiring non-monotonic alignments, such as simultaneous translation. In this research, we address this issue by integrating Transducer's decoding with the history of input stream via a learnable monotonic attention. Our approach leverages the forward-backward algorithm to infer the posterior probability of alignments between the predictor states and input timestamps, which is then used to estimate the monotonic context representations, thereby avoiding the need to enumerate the exponentially large alignment space during training. Extensive experiments show that our MonoAttn-Transducer effectively handles non-monotonic alignments in streaming scenarios, offering a robust solution for complex generation tasks. Code is available at https://github.com/ictnlp/MonoAttn-Transducer.

Scaling up inference-time compute has proven to be a valuable strategy in improving the performance of Large Language Models (LLMs) without fine-tuning. An important task that can benefit from additional inference-time compute is self-repair; given an initial flawed response or guess, the LLM corrects its own mistake and produces an improved response or fix. We leverage the in-context learning ability of LLMs to perform self-repair in the coding domain. The key contribution of our paper is an approach that synthesises and selects an ordered set of golden example pairs, or AuPairs, of these initial guesses and subsequent fixes for the corresponding problems. Each such AuPair is provided as a single in-context example at inference time to generate a repaired solution. For an inference-time compute budget of $N$ LLM calls per problem, $N$ AuPairs are used to generate $N$ repaired solutions, out of which the highest-scoring solution is the final answer. The underlying intuition is that if the LLM is given a different example of fixing an incorrect guess each time, it can subsequently generate a diverse set of repaired solutions. Our algorithm selects these AuPairs in a manner that maximises complementarity and usefulness. We demonstrate the results of our algorithm on 5 LLMs across 7 competitive programming datasets for the code repair task. Our algorithm yields a significant boost in performance compared to best-of-$N$ and self-repair, and also exhibits strong generalisation across datasets and models. Moreover, our approach shows stronger scaling with inference-time compute budget compared to baselines.

Mass spectrometry plays a fundamental role in elucidating the structures of unknown molecules and subsequent scientific discoveries. One formulation of the structure elucidation task is the conditional *de novo* generation of molecular structure given a mass spectrum. Toward a more accurate and efficient scientific discovery pipeline for small molecules, we present DiffMS, a formula-restricted encoder-decoder generative network that achieves state-of-the-art performance on this task. The encoder utilizes a transformer architecture and models mass spectra domain knowledge such as peak formulae and neutral losses, and the decoder is a discrete graph diffusion model restricted by the heavy-atom composition of a known chemical formula. To develop a robust decoder that bridges latent embeddings and molecular structures, we pretrain the diffusion decoder with fingerprint-structure pairs, which are available in virtually infinite quantities, compared to structure-spectrum pairs that number in the tens of thousands. Extensive experiments on established benchmarks show that DiffMS outperforms existing models on *de novo* molecule generation. We provide several ablations to demonstrate the effectiveness of our diffusion and pretraining approaches and show consistent performance scaling with increasing pretraining dataset size. DiffMS code is publicly available at https://github.com/coleygroup/DiffMS.

Click-Through Rate (CTR) prediction models estimate the probability of users clicking on items based on feature interactions, inherently following a discriminative paradigm. However, this paradigm is prone to embedding dimensional collapse and information redundancy due to limitations of vanilla feature embeddings. This motivates us to reformulate it into a generative paradigm to generate new feature embeddings. Unlike sequential recommendation, which naturally fits a generative "next-item prediction" paradigm, it's hard to formulate CTR models into this paradigm without explicit feature order. Therefore, we propose a novel Supervised Feature Generation framework for CTR models, shifting from the discriminative "feature interaction" paradigm to the generative "feature generation" paradigm. Specifically, we predict each feature embedding based on the concatenation of all feature embeddings. Besides, this paradigm naturally accommodates a supervised binary cross-entropy loss to indicate whether the sample is positive or negative. The framework can reformulate nearly every existing CTR model and bring significant performance lifts. Moreover, it produces less-collapsed and redundancy-reduced feature embeddings, thereby mitigating the inherent limitations of the discriminative paradigm. The code can be found at https://github.com/USTC-StarTeam/GE4Rec.

Many commercial Large Language Models (LLMs) are often closed-source, limiting developers to prompt tuning for aligning content generation with specific applications. While these models currently do not provide access to token logits, we argue that if such access were available, it would enable more powerful adaptation techniques beyond prompt engineering. In this paper, we propose a token-level probability reweighting framework that, given access to logits and a small amount of task-specific data, can effectively steer black-box LLMs toward application-specific content generation. Our approach views next-token prediction through the lens of supervised classification. We show that aligning black-box LLMs with task-specific data can be formulated as a label noise correction problem, leading to Plugin model -- an autoregressive probability reweighting model that operates solely on logits. We provide theoretical justification for why reweighting logits alone is sufficient for task adaptation. Extensive experiments with multiple datasets, LLMs, and reweighting models demonstrate the effectiveness of our method, advocating for broader access to token logits in closed-source models. We provide our code at this https URL.