Training large language models (LLMs) with open-domain instruction data has yielded remarkable success in aligning to end tasks and user preferences. Extensive research has highlighted that enhancing the quality and diversity of instruction data consistently improves performance. However, the impact of data complexity, as a crucial metric, remains relatively unexplored in three aspects: (1) scaling law, where the sustainability of performance improvements with increasing complexity is uncertain, (2) additional tokens, whether the improvement brought by complexity comes from introducing more training tokens, and (3) curriculum tuning, where the potential advantages of incorporating instructions ranging from easy to difficult are not yet fully understood. In this paper, we propose \textit{tree-instruct} to systematically enhance the complexity of instruction data in a controllable manner. This approach adds a specified number of nodes into the instruction semantic tree, yielding new instruction data based on the modified tree. By adjusting the number of added nodes, we can control the difficulty level in the modified instruction data. Our preliminary experiments reveal the following insights: (1) Increasing complexity consistently leads to sustained performance improvements. For instance, using 1,000 instruction data and 10 nodes resulted in a substantial 24\% increase in win rate. (2) Under the same token budget, a few complex instructions outperform diverse yet simple instructions. (3) Curriculum instruction tuning might not yield the anticipated results; focusing on increasing complexity appears to be the key.
Measuring the quality of responses generated by LLMs is a challenging task, particularly when it comes to evaluating whether the response is aligned with human preference. A novel approach involves using the LLM itself to make evaluation and stabilizing the results through multiple independent evaluations, similar to a single-layer narrow LLM network. This network consists of a fixed number of neurons, with each neuron being the same LLM. In this paper, we draw upon the extensive research on deep neural networks to explore whether deeper and wider networks can lead to fairer evaluations. Specifically, inspired by the observation that different neurons in a neural network are responsible for detecting different concepts, we first adaptively generate as many neuron roles as possible for each evaluation sample. Each perspective corresponds to the role of a specific LLM neuron in the first layer. In subsequent layers, we follow the idea that higher layers in deep networks are responsible for more comprehensive features, each layer receives representations from all neurons in the previous layer, integrating the locally learned evaluation information to obtain a more comprehensive evaluation result. Interestingly, this network design resembles the process of academic paper reviewing. To validate the effectiveness of our method, we construct the largest and most diverse English evaluation benchmark LLMEval$^2$ for LLM evaluators, comprising 15 tasks, 8 abilities, and 2,553 samples. Experimental results demonstrate that a wider network (involving many reviewers) with 2 layers (one round of discussion) performs the best, improving kappa correlation coefficient from 0.28 to 0.34. We also leverage WideDeep to aid in the assessment of Chinese LLMs, which has accelerated the evaluation time by 4.6 times, resulting in a 60% cost saving. WideDeep achieves a remarkable 93% agreement level among humans.
Large language models (LLMs) demonstrate remarkable ability to comprehend, reason, and generate following nature language instructions. However, the development of LLMs has been primarily focused on high-resource languages, such as English, thereby limiting their applicability and research in other languages. Consequently, we present PolyLM, a multilingual LLM trained on 640 billion (B) tokens, avaliable in two model sizes: 1.7B and 13B. To enhance its multilingual capabilities, we 1) integrate bilingual data into training data; and 2) adopt a curriculum learning strategy that increases the proportion of non-English data from 30% in the first stage to 60% in the final stage during pre-training. Further, we propose a multilingual self-instruct method which automatically generates 132.7K diverse multilingual instructions for model fine-tuning. To assess the model's performance, we collect several existing multilingual tasks, including multilingual understanding, question answering, generation, and translation. Extensive experiments show that PolyLM surpasses other open-source models such as LLaMA and BLOOM on multilingual tasks while maintaining comparable performance in English. Our models, alone with the instruction data and multilingual benchmark, are available at: \url{https://modelscope.cn/models/damo/nlp_polylm_13b_text_generation}.
Large language models (LLMs) often contain misleading content, emphasizing the need to align them with human values to ensure secur AI systems. Reinforcement learning from human feedback (RLHF) has been employed to achieve this alignment by combining a reward model, typically based on Bradley-Terry paired comparison, with an RL algorithm such as Proximal Policy Optimization (PPO) to optimize LLM responses. However, RLHF exhibits complexity, instability, and sensitivity to hyperparameters. In this paper, we propose Preference Ranking Optimization (PRO) as an alternative to PPO for directly aligning LLMs with the Bradley-Terry comparison. PRO extends the pairwise Bradley-Terry comparison to accommodate preference rankings of any length. By iteratively contrasting the likelihood of generating responses, PRO instructs the LLM to prioritize the best response while progressively ranking the remaining responses. In this manner, PRO effectively transforms human alignment into aligning the probability ranking of $n$ responses generated by LLM with the preference ranking of humans towards these responses. Experiments have shown that PRO outperforms existing alignment algorithms, achieving comparable results to ChatGPT and human responses through automatic-based, reward-based, GPT-4, and human evaluations. Furthermore, we demonstrate that longer, more diverse, and higher-quality preference ranking sequences can consistently enhance the performance of human alignment.
When trying to answer complex questions, people often rely on multiple sources of information, such as visual, textual, and tabular data. Previous approaches to this problem have focused on designing input features or model structure in the multi-modal space, which is inflexible for cross-modal reasoning or data-efficient training. In this paper, we call for an alternative paradigm, which transforms the images and tables into unified language representations, so that we can simplify the task into a simpler textual QA problem that can be solved using three steps: retrieval, ranking, and generation, all within a language space. This idea takes advantage of the power of pre-trained language models and is implemented in a framework called Solar. Our experimental results show that Solar outperforms all existing methods by 10.6-32.3 pts on two datasets, MultimodalQA and MMCoQA, across ten different metrics. Additionally, Solar achieves the best performance on the WebQA leaderboard
The goal of document-grounded dialogue (DocGD) is to generate a response by grounding the evidence in a supporting document in accordance with the dialogue context. This process involves four variables that are causally connected. Recently, task-specific pre-training has greatly boosted performances on many downstream tasks. Existing DocGD methods, however, continue to rely on general pre-trained language models without a specifically tailored pre-training approach that explicitly captures the causal relationships. To tackle this issue, we are the first to present a causally-complete dataset construction strategy for building million-level DocGD pre-training corpora. To better capture causality, we further propose a causally-perturbed pre-training strategy, which introduces causal perturbations on the variables and optimizes the overall causal effect. Experiments on three benchmark datasets demonstrate that our causal pre-training achieves considerable and consistent improvements under fully-supervised, low-resource, few-shot, and zero-shot settings.
Lifelong learning (LL) is an important ability for NLP models to learn new tasks continuously. Architecture-based approaches are reported to be effective implementations for LL models. However, it is non-trivial to extend previous approaches to domain incremental LL scenarios since they either require access to task identities in the testing phase or cannot handle samples from unseen tasks. In this paper, we propose \textbf{Diana}: a \underline{d}ynam\underline{i}c \underline{a}rchitecture-based lifelo\underline{n}g le\underline{a}rning model that tries to learn a sequence of tasks with a prompt-enhanced language model. Four types of hierarchically organized prompts are used in Diana to capture knowledge from different granularities. Specifically, we dedicate task-level prompts to capture task-specific knowledge to retain high LL performances and maintain instance-level prompts to learn knowledge shared across input samples to improve the model's generalization performance. Moreover, we dedicate separate prompts to explicitly model unseen tasks and introduce a set of prompt key vectors to facilitate knowledge sharing between tasks. Extensive experiments demonstrate that Diana outperforms state-of-the-art LL models, especially in handling unseen tasks. We release the code and data at \url{https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/diana}.
Recently developed graph contrastive learning (GCL) approaches compare two different "views" of the same graph in order to learn node/graph representations. The core assumption of these approaches is that by graph augmentation, it is possible to generate several structurally different but semantically similar graph structures, and therefore, the identity labels of the original and augmented graph/nodes should be identical. However, in this paper, we observe that this assumption does not always hold, for example, any perturbation to nodes or edges in a molecular graph will change the graph labels to some degree. Therefore, we believe that augmenting the graph structure should be accompanied by an adaptation of the labels used for the contrastive loss. Based on this idea, we propose ID-MixGCL, which allows for simultaneous modulation of both the input graph and the corresponding identity labels, with a controllable degree of change, leading to the capture of fine-grained representations from unlabeled graphs. Experimental results demonstrate that ID-MixGCL improves performance on graph classification and node classification tasks, as demonstrated by significant improvements on the Cora, IMDB-B, and IMDB-M datasets compared to state-of-the-art techniques, by 3-29% absolute points.
Recent research has shown that Large Language Models (LLMs) can utilize external tools to improve their contextual processing abilities, moving away from the pure language modeling paradigm and paving the way for Artificial General Intelligence. Despite this, there has been a lack of systematic evaluation to demonstrate the efficacy of LLMs using tools to respond to human instructions. This paper presents API-Bank, the first benchmark tailored for Tool-Augmented LLMs. API-Bank includes 53 commonly used API tools, a complete Tool-Augmented LLM workflow, and 264 annotated dialogues that encompass a total of 568 API calls. These resources have been designed to thoroughly evaluate LLMs' ability to plan step-by-step API calls, retrieve relevant APIs, and correctly execute API calls to meet human needs. The experimental results show that GPT-3.5 emerges the ability to use the tools relative to GPT3, while GPT-4 has stronger planning performance. Nevertheless, there remains considerable scope for further improvement when compared to human performance. Additionally, detailed error analysis and case studies demonstrate the feasibility of Tool-Augmented LLMs for daily use, as well as the primary challenges that future research needs to address.
Open Information Extraction (OpenIE) facilitates the open-domain discovery of textual facts. However, the prevailing solutions evaluate OpenIE models on in-domain test sets aside from the training corpus, which certainly violates the initial task principle of domain-independence. In this paper, we propose to advance OpenIE towards a more realistic scenario: generalizing over unseen target domains with different data distributions from the source training domains, termed Generalized OpenIE. For this purpose, we first introduce GLOBE, a large-scale human-annotated multi-domain OpenIE benchmark, to examine the robustness of recent OpenIE models to domain shifts, and the relative performance degradation of up to 70% implies the challenges of generalized OpenIE. Then, we propose DragonIE, which explores a minimalist graph expression of textual fact: directed acyclic graph, to improve the OpenIE generalization. Extensive experiments demonstrate that DragonIE beats the previous methods in both in-domain and out-of-domain settings by as much as 6.0% in F1 score absolutely, but there is still ample room for improvement.