LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback

To democratize large language models (LLMs) to most natural languages, it is imperative to make these models capable of understanding and generating texts in many languages, in particular low-resource ones. While recent multilingual LLMs demonstrate remarkable performance in such capabilities, these LLMs still support a limited number of human languages due to the lack of training data for low-resource languages. Moreover, these LLMs are not yet aligned with human preference for downstream tasks, which is crucial for the success of LLMs in English. In this paper, we introduce xLLaMA-100 and xBLOOM-100 (collectively xLLMs-100), which scale the multilingual capabilities of LLaMA and BLOOM to 100 languages. To do so, we construct two datasets: a multilingual instruction dataset including 100 languages, which represents the largest language coverage to date, and a cross-lingual human feedback dataset encompassing 30 languages. We perform multilingual instruction tuning on the constructed instruction data and further align the LLMs with human feedback using the DPO algorithm on our cross-lingual human feedback dataset. We evaluate the multilingual understanding and generating capabilities of xLLMs-100 on five multilingual benchmarks. Experimental results show that xLLMs-100 consistently outperforms its peers across the benchmarks by considerable margins, defining a new state-of-the-art multilingual LLM that supports 100 languages.

FREB-TQA: A Fine-Grained Robustness Evaluation Benchmark for Table Question Answering

Table Question Answering (TQA) aims at composing an answer to a question based on tabular data. While prior research has shown that TQA models lack robustness, understanding the underlying cause and nature of this issue remains predominantly unclear, posing a significant obstacle to the development of robust TQA systems. In this paper, we formalize three major desiderata for a fine-grained evaluation of robustness of TQA systems. They should (i) answer questions regardless of alterations in table structure, (ii) base their responses on the content of relevant cells rather than on biases, and (iii) demonstrate robust numerical reasoning capabilities. To investigate these aspects, we create and publish a novel TQA evaluation benchmark in English. Our extensive experimental analysis reveals that none of the examined state-of-the-art TQA systems consistently excels in these three aspects. Our benchmark is a crucial instrument for monitoring the behavior of TQA systems and paves the way for the development of robust TQA systems. We release our benchmark publicly.

Asking Clarification Questions for Code Generation in General-Purpose Programming Language

Code generation from text requires understanding the user's intent from a natural language description (NLD) and generating an executable program code snippet that satisfies this intent. While recent pretrained language models (PLMs) demonstrate remarkable performance for this task, these models fail when the given NLD is ambiguous due to the lack of enough specifications for generating a high-quality code snippet. In this work, we introduce a novel and more realistic setup for this task. We hypothesize that ambiguities in the specifications of an NLD are resolved by asking clarification questions (CQs). Therefore, we collect and introduce a new dataset named CodeClarQA containing NLD-Code pairs with created CQAs. We evaluate the performance of PLMs for code generation on our dataset. The empirical results support our hypothesis that clarifications result in more precise generated code, as shown by an improvement of 17.52 in BLEU, 12.72 in CodeBLEU, and 7.7\% in the exact match. Alongside this, our task and dataset introduce new challenges to the community, including when and what CQs should be asked.

The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification

Few-shot Intent Classification (FSIC) is one of the key challenges in modular task-oriented dialog systems. While advanced FSIC methods are similar in using pretrained language models to encode texts and nearest neighbour-based inference for classification, these methods differ in details. They start from different pretrained text encoders, use different encoding architectures with varying similarity functions, and adopt different training regimes. Coupling these mostly independent design decisions and the lack of accompanying ablation studies are big obstacle to identify the factors that drive the reported FSIC performance. We study these details across three key dimensions: (1) Encoding architectures: Cross-Encoder vs Bi-Encoders; (2) Similarity function: Parameterized (i.e., trainable) functions vs non-parameterized function; (3) Training regimes: Episodic meta-learning vs the straightforward (i.e., non-episodic) training. Our experimental results on seven FSIC benchmarks reveal three important findings. First, the unexplored combination of the cross-encoder architecture (with parameterized similarity scoring function) and episodic meta-learning consistently yields the best FSIC performance. Second, Episodic training yields a more robust FSIC classifier than non-episodic one. Third, in meta-learning methods, splitting an episode to support and query sets is not a must. Our findings paves the way for conducting state-of-the-art research in FSIC and more importantly raise the community's attention to details of FSIC methods. We release our code and data publicly.

ArgSciChat: A Dataset for Argumentative Dialogues on Scientific Papers

The applications of conversational agents for scientific disciplines (as expert domains) are understudied due to the lack of dialogue data to train such agents. While most data collection frameworks, such as Amazon Mechanical Turk, foster data collection for generic domains by connecting crowd workers and task designers, these frameworks are not much optimized for data collection in expert domains. Scientists are rarely present in these frameworks due to their limited time budget. Therefore, we introduce a novel framework to collect dialogues between scientists as domain experts on scientific papers. Our framework lets scientists present their scientific papers as groundings for dialogues and participate in dialogue they like its paper title. We use our framework to collect a novel argumentative dialogue dataset, ArgSciChat. It consists of 498 messages collected from 41 dialogues on 20 scientific papers. Alongside extensive analysis on ArgSciChat, we evaluate a recent conversational agent on our dataset. Experimental results show that this agent poorly performs on ArgSciChat, motivating further research on argumentative scientific agents. We release our framework and the dataset.

Generating Persona-Consistent Dialogue Responses Using Deep Reinforcement Learning

Recent transformer-based open-domain dialogue agents are trained by reference responses in a fully supervised scenario. Such agents often display inconsistent personalities as training data potentially contain contradictory responses to identical input utterances and no persona-relevant criteria are used in their training losses. We propose a novel approach to train transformer-based dialogue agents using actor-critic reinforcement learning. We define a new reward function to assess generated responses in terms of persona consistency, topic consistency, and fluency. Our reference-agnostic reward relies only on a dialogue history and a persona defined by a list of facts. Automatic and human evaluations on the PERSONACHAT dataset show that our proposed approach increases the rate of persona-consistent responses compared with its peers that are trained in a fully supervised scenario using reference responses.

When is ACL's Deadline? A Scientific Conversational Agent

Our conversational agent UKP-ATHENA assists NLP researchers in finding and exploring scientific literature, identifying relevant authors, planning or post-processing conference visits, and preparing paper submissions using a unified interface based on natural language inputs and responses. UKP-ATHENA enables new access paths to our swiftly evolving research area with its massive amounts of scientific information and high turnaround times. UKP-ATHENA's responses connect information from multiple heterogeneous sources which researchers currently have to explore manually one after another. Unlike a search engine, UKP-ATHENA maintains the context of a conversation to allow for efficient information access on papers, researchers, and conferences. Our architecture consists of multiple components with reference implementations that can be easily extended by new skills and domains. Our user-based evaluation shows that UKP-ATHENA already responds 45% of different formulations of defined intents with 37% information coverage rate.

A Neural Model for Dialogue Coherence Assessment

Dialogue quality assessment is crucial for evaluating dialogue agents. An essential factor of high-quality dialogues is coherence - what makes dialogue utterances a whole. This paper proposes a novel dialogue coherence model trained in a hierarchical multi-task learning scenario where coherence assessment is the primary and the high-level task, and dialogue act prediction is the auxiliary and the low-level task. The results of our experiments for two benchmark dialogue corpora (i.e. SwitchBoard and DailyDialog) show that our model significantly outperforms its competitors for ranking dialogues with respect to their coherence. Although the performance of other examined models considerably varies across examined corpora, our model robustly achieves high performance. We release the source code and datasets defined for the experiments in this paper to accelerate future research on dialogue coherence.

Reward Learning for Efficient Reinforcement Learning in Extractive Document Summarisation

Document summarisation can be formulated as a sequential decision-making problem, which can be solved by Reinforcement Learning (RL) algorithms. The predominant RL paradigm for summarisation learns a cross-input policy, which requires considerable time, data and parameter tuning due to the huge search spaces and the delayed rewards. Learning input-specific RL policies is a more efficient alternative but so far depends on handcrafted rewards, which are difficult to design and yield poor performance. We propose RELIS, a novel RL paradigm that learns a reward function with Learning-to-Rank (L2R) algorithms at training time and uses this reward function to train an input-specific RL policy at test time. We prove that RELIS guarantees to generate near-optimal summaries with appropriate L2R and RL algorithms. Empirically, we evaluate our approach on extractive multi-document summarisation. We show that RELIS reduces the training time by two orders of magnitude compared to the state-of-the-art models while performing on par with them.

Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systems

Visual modifications to text are often used to obfuscate offensive comments in social media (e.g., "!d10t") or as a writing style ("1337" in "leet speak"), among other scenarios. We consider this as a new type of adversarial attack in NLP, a setting to which humans are very robust, as our experiments with both simple and more difficult visual input perturbations demonstrate. We then investigate the impact of visual adversarial attacks on current NLP systems on character-, word-, and sentence-level tasks, showing that both neural and non-neural models are, in contrast to humans, extremely sensitive to such attacks, suffering performance decreases of up to 82\%. We then explore three shielding methods---visual character embeddings, adversarial training, and rule-based recovery---which substantially improve the robustness of the models. However, the shielding methods still fall behind performances achieved in non-attack scenarios, which demonstrates the difficulty of dealing with visual attacks.