相关论文: Geometric Analysis of Token Selection in Multi-Hea…

Large Language Models (LLMs) tend to attend heavily to the first token in the sequence -- creating a so-called attention sink. Many works have studied this phenomenon in detail, proposing various ways to either leverage or alleviate it.…

Recent studies have shown that Large Language Models (LLMs) can achieve strong reasoning performance by incorporating functional symbolic representations that abstractly describe graph traversal algorithms and step-by-step reasoning in…

Recent advances in multimodal large language models enable new possibilities for image-based decision support. However, their reliability and operational trade-offs in neuroimaging remain insufficiently understood. We present a…

Fine-tuning Large Language Models (LLMs) is an effective method to enhance their performance on downstream tasks. However, choosing the appropriate setting of tuning hyperparameters (HPs) is a labor-intensive and computationally expensive…

Predictive geometric models deliver excellent results for many Machine Learning use cases. Despite their undoubted performance, neural predictive algorithms can show unexpected degrees of instability and variance, particularly when applied…

Steering, or direct manipulation of internal activations to guide LLM responses toward specific semantic concepts, is emerging as a promising avenue for both understanding how semantic concepts are stored within LLMs and advancing LLM…

We study rigid-body motion planning through multiple sequential narrow openings, which requires long-horizon geometric reasoning because the configuration used to traverse an early opening constrains the set of reachable configurations for…

Large Reasoning Models (LRMs) have shown promising accuracy improvements on complex problem-solving tasks. While these models have attained high accuracy by leveraging additional computation at test time, they need to generate long…

The three-dimensional shape and conformation of small-molecule ligands are critical for biomolecular recognition, yet encoding 3D geometry has not improved ligand-based virtual screening approaches. We describe an end-to-end deep learning…

In this work, we investigate several neural network architectures for fine-grained entity type classification. Particularly, we consider extensions to a recently proposed attentive neural architecture and make three key contributions.…

The geometric evolution of token representations in large language models (LLMs) presents a fundamental paradox: while human language inherently organizes semantic information in low-dimensional spaces ($\sim 10^1$ dimensions), modern LLMs…

Multimodal large language models (MLLMs) have exhibited remarkable performance in various visual tasks, yet still struggle with spatial reasoning. Recent efforts mitigate this by injecting geometric features from 3D foundation models, but…

Large Language Models (LLMs) have emerged as a pivotal research area, yet the attention module remains a critical bottleneck in LLM inference, even with techniques like KVCache to mitigate redundant computations. While various top-$k$…

This paper investigates how Transformer language models (LMs) fine-tuned for acceptability classification capture linguistic features. Our approach uses the best practices of topological data analysis (TDA) in NLP: we construct directed…

Multimodal Large Language Models (MLLMs) have displayed remarkable performance in multi-modal tasks, particularly in visual comprehension. However, we reveal that MLLMs often generate incorrect answers even when they understand the visual…

Visual token reduction lowers inference costs caused by extensive image features in large vision-language models (LVLMs). Unlike relevant studies that prune tokens in self-attention-only LVLMs, our work uniquely addresses…

Understanding the latent space geometry of large language models (LLMs) is key to interpreting their behavior and improving alignment. Yet it remains unclear to what extent LLMs linearly organize representations related to semantic…

The quality of meeting summaries generated by natural language generation (NLG) systems is hard to measure automatically. Established metrics such as ROUGE and BERTScore have a relatively low correlation with human judgments and fail to…

Reinforcement-learning-based post-training has become a key approach for improving the reasoning ability of large language models, but its token-level learning signals remain poorly understood. This work studies their heterogeneity through…

Grouping has been commonly used in deep metric learning for computing diverse features. However, current methods are prone to overfitting and lack interpretability. In this work, we propose an improved and interpretable grouping method to…