Search

Large language models (LLMs) have advanced financial applications, yet they often lack sufficient financial knowledge and struggle with tasks involving multi-modal inputs like tables and time series data. To address these limitations, we introduce \textit{Open-FinLLMs}, a series of Financial LLMs. We begin with FinLLaMA, pre-trained on a 52 billion token financial corpus, incorporating text, tables, and time-series data to embed comprehensive financial knowledge. FinLLaMA is then instruction fine-tuned with 573K financial instructions, resulting in FinLLaMA-instruct, which enhances task performance. Finally, we present FinLLaVA, a multimodal LLM trained with 1.43M image-text instructions to handle complex financial data types. Extensive evaluations demonstrate FinLLaMA's superior performance over LLaMA3-8B, LLaMA3.1-8B, and BloombergGPT in both zero-shot and few-shot settings across 19 and 4 datasets, respectively. FinLLaMA-instruct outperforms GPT-4 and other Financial LLMs on 15 datasets. FinLLaVA excels in understanding tables and charts across 4 multimodal tasks. Additionally, FinLLaMA achieves impressive Sharpe Ratios in trading simulations, highlighting its robust financial application capabilities. We will continually maintain and improve our models and benchmarks to support ongoing innovation in academia and industry., Comment: 33 pages, 13 figures

Over the past decade, the rapid advancement of deep learning and big data applications has been driven by vast datasets and high-performance computing systems. However, as we approach the physical limits of semiconductor fabrication in the post-Moore's Law era, questions arise about the future of these applications. In parallel, quantum computing has made significant progress with the potential to break limits. Major companies like IBM, Google, and Microsoft provide access to noisy intermediate-scale quantum (NISQ) computers. Despite the theoretical promise of Shor's and Grover's algorithms, practical implementation on current quantum devices faces challenges, such as demanding additional resources and a high number of controlled operations. To tackle these challenges and optimize the utilization of limited onboard qubits, we introduce ReSaQuS, a resource-efficient index-value searching system within a quantum-classical hybrid framework. Building on Grover's algorithm, ReSaQuS employs an automatically managed iterative search approach. This method analyzes problem size, filters fewer probable data points, and progressively reduces the dataset with decreasing qubit requirements. Implemented using Qiskit and evaluated through extensive experiments, ReSaQuS has demonstrated a substantial reduction, up to 86.36\% in cumulative qubit consumption and 72.72\% in active periods, reinforcing its potential in optimizing quantum computing application deployment.

Temporal Knowledge Graph Reasoning (TKGR) is the task of inferring missing facts for incomplete TKGs in complex scenarios (e.g., transductive and inductive settings), which has been gaining increasing attention. Recently, to mitigate dependence on structured connections in TKGs, text-based methods have been developed to utilize rich linguistic information from entity descriptions. However, suffering from the enormous parameters and inflexibility of pre-trained language models, existing text-based methods struggle to balance the textual knowledge and temporal information with computationally expensive purpose-built training strategies. To tap the potential of text-based models for TKGR in various complex scenarios, we propose ChapTER, a Contrastive historical modeling framework with prefix-tuning for TEmporal Reasoning. ChapTER feeds history-contextualized text into the pseudo-Siamese encoders to strike a textual-temporal balance via contrastive estimation between queries and candidates. By introducing virtual time prefix tokens, it applies a prefix-based tuning method to facilitate the frozen PLM capable for TKGR tasks under different settings. We evaluate ChapTER on four transductive and three few-shot inductive TKGR benchmarks, and experimental results demonstrate that ChapTER achieves superior performance compared to competitive baselines with only 0.17% tuned parameters. We conduct thorough analysis to verify the effectiveness, flexibility and efficiency of ChapTER., Comment: Accepted to NAACL 2024 Findings

Multimodal large language models are typically trained in two stages: first pre-training on image-text pairs, and then fine-tuning using supervised vision-language instruction data. Recent studies have shown that large language models can achieve satisfactory results even with a limited amount of high-quality instruction-following data. In this paper, we introduce InstructionGPT-4, which is fine-tuned on a small dataset comprising only 200 examples, amounting to approximately 6\% of the instruction-following data used in the alignment dataset for MiniGPT-4. To achieve this, we first propose several metrics to access the quality of multimodal instruction data. Based on these metrics, we present an effective and trainable data selector to automatically identify and filter low-quality vision-language data. By employing this method, InstructionGPT-4 outperforms the original MiniGPT-4 on various evaluations. Overall, our findings demonstrate that less but high-quality instruction tuning data is efficient in enabling multimodal large language models to generate better output. Our code is available at https://github.com/waltonfuture/InstructionGPT-4.

Few-shot learning aims to train models that can be generalized to novel classes with only a few samples. Recently, a line of works are proposed to enhance few-shot learning with accessible semantic information from class names. However, these works focus on improving existing modules such as visual prototypes and feature extractors of the standard few-shot learning framework. This limits the full potential use of semantic information. In this paper, we propose a novel few-shot learning framework that uses pre-trained language models based on contrastive learning. To address the challenge of alignment between visual features and textual embeddings obtained from text-based pre-trained language model, we carefully design the textual branch of our framework and introduce a metric module to generalize the cosine similarity. For better transferability, we let the metric module adapt to different few-shot tasks and adopt MAML to train the model via bi-level optimization. Moreover, we conduct extensive experiments on multiple benchmarks to demonstrate the effectiveness of our method.

Recently, self-supervised learning has attracted great attention, since it only requires unlabeled data for model training. Contrastive learning is one popular method for self-supervised learning and has achieved promising empirical performance. However, the theoretical understanding of its generalization ability is still limited. To this end, we define a kind of $(\sigma,\delta)$-measure to mathematically quantify the data augmentation, and then provide an upper bound of the downstream classification error rate based on the measure. It reveals that the generalization ability of contrastive self-supervised learning is related to three key factors: alignment of positive samples, divergence of class centers, and concentration of augmented data. The first two factors are properties of learned representations, while the third one is determined by pre-defined data augmentation. We further investigate two canonical contrastive losses, InfoNCE and cross-correlation, to show how they provably achieve the first two factors. Moreover, we conduct experiments to study the third factor, and observe a strong correlation between downstream performance and the concentration of augmented data., Comment: Accepted by ICLR 2023

We study the behavior of eight diffuse interstellar bands (DIBs) in different interstellar environments, as characterized by the fraction of hydrogen in molecular form [$f$(H$_2$)], with comparisons to the corresponding behavior of various known atomic and molecular species. The equivalent widths of the five "normal" DIBs ($\lambda\lambda$5780.5, 5797.1, 6196.0, 6283.8, and 6613.6), normalized to $E(B-V)$, show a "Lambda-shaped" behavior: they increase at low $f$(H$_2$), peak at $f$(H$_2$) ~ 0.3, and then decrease. The similarly normalized column densities of Ca, Ca$^+$, Ti$^+$, and CH$^+$ also decline for $f$(H$_2$) > 0.3. In contrast, the normalized column densities of Na, K, CH, CN, and CO increase monotonically with $f$(H$_2$), and the trends exhibited by the three C$_2$ DIBs ($\lambda\lambda$4726.8, 4963.9, and 4984.8) lie between those two general behaviors. These trends with $f$(H$_2$) are accompanied by cosmic scatter, the dispersion at any given $f$(H$_2$) being significantly larger than the individual errors of measurement. The Lambda-shaped trends suggest the balance between creation and destruction of the DIB carriers differs dramatically between diffuse atomic and diffuse molecular clouds; additional processes besides ionization and shielding are needed to explain those observed trends. Except for several special cases, the highest $W$(5780)/$W$(5797) ratios, characterizing the so-called "sigma-zeta effect", occur only at $f$(H$_2$) < 0.2. We propose a sequence of DIBs based on trends in their pair-wise strength ratios with increasing $f$(H$_2$). In order of increasing environmental density, we find the $\lambda$6283.8 and $\lambda$5780.5 DIBs, the $\lambda$6196.0 DIB, the $\lambda$6613.6 DIB, the $\lambda$5797.1 DIB, and the C$_2$ DIBs., Comment: accepted to ApJ; 53 pages; 12 figures; 3 tables