Selective Attention: Enhancing Transformer through Principled Context Control

The attention mechanism within the transformer architecture enables the model to weigh and combine tokens based on their relevance to the query. While self-attention has enjoyed major success, it notably treats all queries $q$ in the same way by applying the mapping $V^\top\text{softmax}(Kq)$, where $V,K$ are the value and key embeddings respectively. In this work, we argue that this uniform treatment hinders the ability to control contextual sparsity and relevance. As a solution, we introduce the $\textit{Selective Self-Attention}$ (SSA) layer that augments the softmax nonlinearity with a principled temperature scaling strategy. By controlling temperature, SSA adapts the contextual sparsity of the attention map to the query embedding and its position in the context window. Through theory and experiments, we demonstrate that this alleviates attention dilution, aids the optimization process, and enhances the model's ability to control softmax spikiness of individual queries. We also incorporate temperature scaling for value embeddings and show that it boosts the model's ability to suppress irrelevant/noisy tokens. Notably, SSA is a lightweight method which introduces less than 0.5% new parameters through a weight-sharing strategy and can be fine-tuned on existing LLMs. Extensive empirical evaluations demonstrate that SSA-equipped models achieve a noticeable and consistent accuracy improvement on language modeling benchmarks.

CPE-Pro: A Structure-Sensitive Deep Learning Model for Protein Representation and Origin Evaluation

Protein structures are important for understanding their functions and interactions. Currently, many protein structure prediction methods are enriching the structure database. Discriminating the origin of structures is crucial for distinguishing between experimentally resolved and computationally predicted structures, evaluating the reliability of prediction methods, and guiding downstream biological studies. Building on works in structure prediction, We developed a structure-sensitive supervised deep learning model, Crystal vs Predicted Evaluator for Protein Structure (CPE-Pro), to represent and discriminate the origin of protein structures. CPE-Pro learns the structural information of proteins and captures inter-structural differences to achieve accurate traceability on four data classes, and is expected to be extended to more. Simultaneously, we utilized Foldseek to encode protein structures into "structure-sequence" and trained a protein Structural Sequence Language Model, SSLM. Preliminary experiments demonstrated that, compared to large-scale protein language models pre-trained on vast amounts of amino acid sequences, the "structure-sequences" enable the language model to learn more informative protein features, enhancing and optimizing structural representations. We have provided the code, model weights, and all related materials on https://github.com/GouWenrui/CPE-Pro-main.git.

RiTeK: A Dataset for Large Language Models Complex Reasoning over Textual Knowledge Graphs

Answering complex real-world questions often requires accurate retrieval from textual knowledge graphs (TKGs). The scarcity of annotated data, along with intricate topological structures, makes this task particularly challenging. As the nature of relational path information could enhance the inference ability of Large Language Models (LLMs), efficiently retrieving more complex relational path information from TKGs presents another key challenge. To tackle these challenges, we first develop a Dataset for LLMs Complex Reasoning over Textual Knowledge Graphs (RiTeK) with a broad topological structure coverage.We synthesize realistic user queries that integrate diverse topological structures, relational information, and complex textual descriptions. We conduct rigorous expert evaluation to validate the quality of our synthesized queries. And then, we introduce an enhanced Monte Carlo Tree Search (MCTS) method, Relational MCTS, to automatically extract relational path information from textual graphs for specific queries. Our dataset mainly covers the medical domain as the relation types and entity are complex and publicly available. Experimental results indicate that RiTeK poses significant challenges for current retrieval and LLM systems, while the proposed Relational MCTS method enhances LLM inference ability and achieves state-of-the-art performance on RiTeK.

On the Power of Convolution Augmented Transformer

The transformer architecture has catalyzed revolutionary advances in language modeling. However, recent architectural recipes, such as state-space models, have bridged the performance gap. Motivated by this, we examine the benefits of Convolution-Augmented Transformer (CAT) for recall, copying, and length generalization tasks. CAT incorporates convolutional filters in the K/Q/V embeddings of an attention layer. Through CAT, we show that the locality of the convolution synergizes with the global view of the attention. Unlike comparable architectures, such as Mamba or transformer, CAT can provably solve the associative recall (AR) and copying tasks using a single layer while also enjoying guaranteed length generalization. We also establish computational tradeoffs between convolution and attention by characterizing how convolution can mitigate the need for full attention by summarizing the context window and creating salient summary tokens to attend. Evaluations on real datasets corroborate our findings and demonstrate that CAT and its variations indeed enhance the language modeling performance.

CancerLLM: A Large Language Model in Cancer Domain

Medical Large Language Models (LLMs) such as ClinicalCamel 70B, Llama3-OpenBioLLM 70B have demonstrated impressive performance on a wide variety of medical NLP task.However, there still lacks a large language model (LLM) specifically designed for cancer domain. Moreover, these LLMs typically have billions of parameters, making them computationally expensive for healthcare systems.Thus, in this study, we propose CancerLLM, a model with 7 billion parameters and a Mistral-style architecture, pre-trained on 2,676,642 clinical notes and 515,524 pathology reports covering 17 cancer types, followed by fine-tuning on three cancer-relevant tasks, including cancer phenotypes extraction, cancer diagnosis generation, and cancer treatment plan generation. Our evaluation demonstrated that CancerLLM achieves state-of-the-art results compared to other existing LLMs, with an average F1 score improvement of 8.1\%. Additionally, CancerLLM outperforms other models on two proposed robustness testbeds. This illustrates that CancerLLM can be effectively applied to clinical AI systems, enhancing clinical research and healthcare delivery in the field of cancer.

Benchmarking Retrieval-Augmented Large Language Models in Biomedical NLP: Application, Robustness, and Self-Awareness

Large language models (LLM) have demonstrated remarkable capabilities in various biomedical natural language processing (NLP) tasks, leveraging the demonstration within the input context to adapt to new tasks. However, LLM is sensitive to the selection of demonstrations. To address the hallucination issue inherent in LLM, retrieval-augmented LLM (RAL) offers a solution by retrieving pertinent information from an established database. Nonetheless, existing research work lacks rigorous evaluation of the impact of retrieval-augmented large language models on different biomedical NLP tasks. This deficiency makes it challenging to ascertain the capabilities of RAL within the biomedical domain. Moreover, the outputs from RAL are affected by retrieving the unlabeled, counterfactual, or diverse knowledge that is not well studied in the biomedical domain. However, such knowledge is common in the real world. Finally, exploring the self-awareness ability is also crucial for the RAL system. So, in this paper, we systematically investigate the impact of RALs on 5 different biomedical tasks (triple extraction, link prediction, classification, question answering, and natural language inference). We analyze the performance of RALs in four fundamental abilities, including unlabeled robustness, counterfactual robustness, diverse robustness, and negative awareness. To this end, we proposed an evaluation framework to assess the RALs' performance on different biomedical NLP tasks and establish four different testbeds based on the aforementioned fundamental abilities. Then, we evaluate 3 representative LLMs with 3 different retrievers on 5 tasks over 9 datasets.

BiomedRAG: A Retrieval Augmented Large Language Model for Biomedicine

Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains; however, these models encounter issues such as generating inaccurate information or hallucinations. Retrieval-augmented generation provided a solution for these models to update knowledge and enhance their performance. In contrast to previous retrieval-augmented LMs, which utilize specialized cross-attention mechanisms to help LLM encode retrieved text, BiomedRAG adopts a simpler approach by directly inputting the retrieved chunk-based documents into the LLM. This straightforward design is easily applicable to existing retrieval and language models, effectively bypassing noise information in retrieved documents, particularly in noise-intensive tasks. Moreover, we demonstrate the potential for utilizing the LLM to supervise the retrieval model in the biomedical domain, enabling it to retrieve the document that assists the LM in improving its predictions. Our experiments reveal that with the tuned scorer,\textsc{ BiomedRAG} attains superior performance across 5 biomedical NLP tasks, encompassing information extraction (triple extraction, relation extraction), text classification, link prediction, and question-answering, leveraging over 9 datasets. For instance, in the triple extraction task, \textsc{BiomedRAG} outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively.

Simple, Efficient and Scalable Structure-aware Adapter Boosts Protein Language Models

Fine-tuning Pre-trained protein language models (PLMs) has emerged as a prominent strategy for enhancing downstream prediction tasks, often outperforming traditional supervised learning approaches. As a widely applied powerful technique in natural language processing, employing Parameter-Efficient Fine-Tuning techniques could potentially enhance the performance of PLMs. However, the direct transfer to life science tasks is non-trivial due to the different training strategies and data forms. To address this gap, we introduce SES-Adapter, a simple, efficient, and scalable adapter method for enhancing the representation learning of PLMs. SES-Adapter incorporates PLM embeddings with structural sequence embeddings to create structure-aware representations. We show that the proposed method is compatible with different PLM architectures and across diverse tasks. Extensive evaluations are conducted on 2 types of folding structures with notable quality differences, 9 state-of-the-art baselines, and 9 benchmark datasets across distinct downstream tasks. Results show that compared to vanilla PLMs, SES-Adapter improves downstream task performance by a maximum of 11% and an average of 3%, with significantly accelerated training speed by a maximum of 1034% and an average of 362%, the convergence rate is also improved by approximately 2 times. Moreover, positive optimization is observed even with low-quality predicted structures. The source code for SES-Adapter is available at https://github.com/tyang816/SES-Adapter.

A Condensed Transition Graph Framework for Zero-shot Link Prediction with Large Language Models

Zero-shot link prediction (ZSLP) on knowledge graphs aims at automatically identifying relations between given entities. Existing methods primarily employ auxiliary information to predict tail entity given head entity and its relation, yet face challenges due to the occasional unavailability of such detailed information and the inherent simplicity of predicting tail entities based on semantic similarities. Even though Large Language Models (LLMs) offer a promising solution to predict unobserved relations between the head and tail entity in a zero-shot manner, their performance is still restricted due to the inability to leverage all the (exponentially many) paths' information between two entities, which are critical in collectively indicating their relation types. To address this, in this work, we introduce a Condensed Transition Graph Framework for Zero-Shot Link Prediction (CTLP), which encodes all the paths' information in linear time complexity to predict unseen relations between entities, attaining both efficiency and information preservation. Specifically, we design a condensed transition graph encoder with theoretical guarantees on its coverage, expressiveness, and efficiency. It is learned by a transition graph contrastive learning strategy. Subsequently, we design a soft instruction tuning to learn and map the all-path embedding to the input of LLMs. Experimental results show that our proposed CTLP method achieves state-of-the-art performance on three standard ZSLP datasets

Class-attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective

Modern classification problems exhibit heterogeneities across individual classes: Each class may have unique attributes, such as sample size, label quality, or predictability (easy vs difficult), and variable importance at test-time. Without care, these heterogeneities impede the learning process, most notably, when optimizing fairness objectives. Confirming this, under a gaussian mixture setting, we show that the optimal SVM classifier for balanced accuracy needs to be adaptive to the class attributes. This motivates us to propose CAP: An effective and general method that generates a class-specific learning strategy (e.g. hyperparameter) based on the attributes of that class. This way, optimization process better adapts to heterogeneities. CAP leads to substantial improvements over the naive approach of assigning separate hyperparameters to each class. We instantiate CAP for loss function design and post-hoc logit adjustment, with emphasis on label-imbalanced problems. We show that CAP is competitive with prior art and its flexibility unlocks clear benefits for fairness objectives beyond balanced accuracy. Finally, we evaluate CAP on problems with label noise as well as weighted test objectives to showcase how CAP can jointly adapt to different heterogeneities.