In recent years, large language models (LLMs) have achieved remarkable success in the field of natural language generation. Compared to previous small-scale models, they are capable of generating fluent output based on the provided prefix or prompt. However, one critical challenge — the *hallucination* problem — remains to be resolved. Generally, the community refers to the undetected hallucination scenario where the LLMs generate text unrelated to the input text or facts. In this study, we intend to model the distributional distance between the regular conditional output and the unconditional output, which is generated without a given input text. Based upon Taylor Expansion for this distance at the output probability space, our approach manages to leverage the embedding and first-order gradient information. The resulting approach is plug-and-play that can be easily adapted to any autoregressive LLM. On the hallucination benchmarks HADES and other datasets, our approach achieves state-of-the-art performance.

We introduce FinDVer, a comprehensive benchmark specifically designed to evaluate the explainable claim verification capabilities of LLMs in the context of understanding and analyzing long, hybrid-content financial documents. FinDVer contains 4,000 expert-annotated examples across four subsets, each focusing on a type of scenario that frequently arises in real-world financial domains. We assess a broad spectrum of 25 LLMs under long-context and RAG settings. Our results show that even the current best-performing system (i.e., GPT-4o) significantly lags behind human experts. Our detailed findings and insights highlight the strengths and limitations of existing LLMs in this new task. We believe FinDVer can serve as a valuable benchmark for evaluating LLM capabilities in claim verification over complex, expert-domain documents.

Prompt recovery in large language models (LLMs) is crucial for understanding how LLMs work and addressing concerns regarding privacy, copyright, etc. The trend towards inference-only APIs complicates this task by restricting access to essential outputs for recovery. To tackle this challenge, we extract prompt-related information from limited outputs and identify a strong(negative) correlation between output probability-based uncertainty and the success of prompt recovery.This finding led to the development of Deliberative PrOmpt RecoverY (DORY), our novel approach that leverages uncertainty to recover prompts accurately. DORY involves reconstructing drafts from outputs, refining these with hints, and filtering out noise based on uncertainty. Our evaluation shows that DORY outperforms existing baselines across diverse LLMs and prompt benchmarks, improving performance by approximately 10.82% and establishing a new state-of-the-art record in prompt recovery tasks. Significantly, DORY operates using a single LLM without any external resources or model, offering a cost-effective, user-friendly prompt recovery solution.

In the current landscape of large language models (LLMs), the process of instruction tuning serves as an essential step. Considering the high computing power overhead, data-efficient instruction tuning was proposed to reduce the training data size in this process, aiming at selecting high-quality instructional data. Nevertheless, we argue that most current data-efficient instruction-tuning methods are highly dependent on the quality of the original instruction-tuning dataset. When it comes to datasets synthesized by LLMs, a common scenario in this field, dirty samples will even be selected with a higher probability than other samples. To address these challenges, we utilized external knowledge (relevant examples or paragraphs) to evaluate those samples synthesized by LLMs with an in-context-based relative predictive entropy. Based on the new metric, we proposed a framework, dubbed as RECOST, which integrates external-knowledge-base re-ranking and diversity-consistent sampling into a single pipeline. Through extensive experiments on several synthetic datasets (Alpaca and Alpaca-gpt4), we demonstrate the effectiveness of our method and achieve even better results with only 1% of the full dataset.

In this paper, we extend financial sentiment analysis (FSA) to event-level since events usually serve as the subject of the sentiment in financial text. Though extracting events from the financial text may be conducive to accurate sentiment predictions, it has specialized challenges due to the lengthy and discontinuity of events in a financial text. To this end, we reconceptualize the event extraction as a classification task by designing a categorization comprising coarse-grained and fine-grained event categories. Under this setting, we formulate the Event-Level Financial Sentiment Analysis(EFSA for short) task that outputs quintuples consisting of (company, industry, coarse-grained event, fine-grained event, sentiment) from financial text. A large-scale Chinese dataset containing 12,160 news articles and 13,725 quintuples is publicized as a brand new testbed for our task. A four-hop Chain-of-Thought LLM-based approach is devised for this task. Systematically investigations are conducted on our dataset, and the empirical results demonstrate the benchmarking scores of existing methods and our proposed method can reach the current state-of-the-art. Our dataset and framework implementation are available at https://github.com/cty1934/EFSA

Large language models (LLMs) have achieved great success in a variety of natural language understanding tasks. However, domain discrepancies between the downstream task and the pre-training corpora may have hurdled LLMs to excel further in the vertical applications. Contrary to prior computational-heavy methods, we propose a lightweight solution to further bridge the gap in applying LLMs to diverse downstream tasks — a Fast Adaptation method for LLMs via Prompted Data, in short FAvPD. Notably, with FAvPD, we establish an additional adaptive tuning procedure, wherein we integrate downstream text corpora, gold labels as well as external knowledge sources and then envelop them into a form of highly controllable prompt. As a simple, easy-to-use, and versatile solution, FAvPD lies in the intersection of regimes like knowledge-augmented LLMs, fine-tuning, and adaptation techniques. With extensive experiments, we prove that FAvPD excels in both performance efficacy and training efficiency over related prior works. FAvPD is publicly available at https://github.com/Hyatio/FAvPD.

Natural language explanations have the potential to provide rich information that in principle guides model reasoning. Yet, recent work by Lampinen et al. has shown limited utility of natural language explanations in improving classification. To effectively learn from explanations, we present FLamE, a two-stage few-shot learning framework that first generates explanations using GPT-3, and then fine-tunes a smaller model (e.g., RoBERTa) with generated explanations. Our experiments on natural language inference demonstrate effectiveness over strong baselines, increasing accuracy by 17.6% over GPT-3 Babbage and 5.7% over GPT-3 Davinci in e-SNLI.Despite improving classification performance, human evaluation surprisingly reveals that the majority of generated explanations does not adequately justify classification decisions. Additional analyses point to the important role of label-specific cues (e.g., “not know” for the neutral label) in generated explanations.

Toxicity annotators and content moderators often default to mental shortcuts when making decisions. This can lead to subtle toxicity being missed, and seemingly toxic but harmless content being over-detected. We introduce BiasX, a framework that enhances content moderation setups with free-text explanations of statements’ implied social biases, and explore its effectiveness through a large-scale crowdsourced user study. We show that indeed, participants substantially benefit from explanations for correctly identifying subtly (non-)toxic content. The quality of explanations is critical: imperfect machine-generated explanations (+2.4% on hard toxic examples) help less compared to expert-written human explanations (+7.2%). Our results showcase the promise of using free-text explanations to encourage more thoughtful toxicity moderation.

In recent years, large language models (LLMs), such as GPTs, have attained great impact worldwide. However, how to adapt these LLMs to better suit the vertical domain-specific tasks by utilizing external knowledge remains not completely solved. Indeed, there have emerged a few works on this line where most of them rely on an alignment heuristic that is built to inject the corresponding knowledge tuple into the associated text sample. However, despite the promise, we identify a pivotal problem in this work ubiquitously. Simply put, we find that injecting unaligned (i.e., random) knowledge tuple into the LLMs achieves comparable (and sometimes better) results than the aligned knowledge being injected. We therefore take a thorough investigation of this frustrating finding on a variety of related prior work and further provide a chain of potential interpretations for the phenomenon. Based on all that, we offer a simple remediated technique. Briefly, the core of this technique roots in an ideological emphasis on the pruning and purification of the external knowledge base to be injected into LLMs. At last, we show that by integrating this technique into most (if not all) knowledge injection frameworks and recent LLMs, it manages to overcome the aforementioned sanity problem and further pushes the boundary of the performance of the domain-adaptive LLMs.

Despite the strong performance of current NLP models, they can be brittle against adversarial attacks. To enable effective learning against adversarial inputs, we introduce the use of rationale models that can explicitly learn to ignore attack tokens. We find that the rationale models can successfully ignore over 90% of attack tokens. This approach leads to consistent sizable improvements (~10%) over baseline models in robustness on three datasets for both BERT and RoBERTa, and also reliably outperforms data augmentation with adversarial examples alone. In many cases, we find that our method is able to close the gap between model performance on a clean test set and an attacked test set and hence reduce the effect of adversarial attacks.

With a handful of demonstration examples, large-scale language models demonstrate strong capability to perform various tasks by in-context learning from these examples, without any fine-tuning. We demonstrate that in-context learning performance can be highly unstable across samples of examples, indicating the idiosyncrasies of how language models acquire information. We formulate example selection for in-context learning as a sequential decision problem, and propose a reinforcement learning algorithm for identifying generalizable policies to select demonstration examples. For GPT-2, our learned policies demonstrate strong abilities of generalizing to unseen tasks in training, with a 5.8% improvement on average. Examples selected from our learned policies can even achieve a small improvement on GPT-3 Ada. However, the improvement diminishes on larger GPT-3 models, suggesting emerging capabilities of large language models.

The aspect-based sentiment analysis (ABSA) is a fine-grained task that aims to determine the sentiment polarity towards targeted aspect terms occurring in the sentence. The development of the ABSA task is very much hindered by the lack of annotated data. To tackle this, the prior works have studied the possibility of utilizing the sentiment analysis (SA) datasets to assist in training the ABSA model, primarily via pretraining or multi-task learning. In this article, we follow this line, and for the first time, we manage to apply the Pseudo-Label (PL) method to merge the two homogeneous tasks. While it seems straightforward to use generated pseudo labels to handle this case of label granularity unification for two highly related tasks, we identify its major challenge in this paper and propose a novel framework, dubbed as Dual-granularity Pseudo Labeling (DPL). Further, similar to PL, we regard the DPL as a general framework capable of combining other prior methods in the literature. Through extensive experiments, DPL has achieved state-of-the-art performance on standard benchmarks surpassing the prior work significantly.

In this paper, we introduce a simple system Baidu submitted for MRQA (Machine Reading for Question Answering) 2019 Shared Task that focused on generalization of machine reading comprehension (MRC) models. Our system is built on a framework of pretraining and fine-tuning, namely D-NET. The techniques of pre-trained language models and multi-task learning are explored to improve the generalization of MRC models and we conduct experiments to examine the effectiveness of these strategies. Our system is ranked at top 1 of all the participants in terms of averaged F1 score. Our codes and models will be released at PaddleNLP.