Humanitarian organizations can enhance their effectiveness by analyzing data to discover trends, gather aggregated insights, manage their security risks, support decision-making, and inform advocacy and funding proposals. However, data about violent incidents with direct impact and relevance for humanitarian aid operations is not readily available. An automatic data collection and NLP-backed classification framework aligned with humanitarian perspectives can help bridge this gap. In this paper, we present HumVI – a dataset comprising news articles in three languages (English, French, Arabic) containing instances of different types of violent incidents categorized by the humanitarian sector they impact, e.g., aid security, education, food security, health, and protection. Reliable labels were obtained for the dataset by partnering with a data-backed humanitarian organization, Insecurity Insight. We provide multiple benchmarks for the dataset, employing various deep learning architectures and techniques, including data augmentation and mask loss, to address different task-related challenges, e.g., domain expansion. The dataset is publicly available at https://github.com/dataminr-ai/humvi-dataset.
论文检索
输入标题、作者或关键词，从 2,271 篇学术成果中精准定位
Long-context large language models (LLMs) hold promise for tasks such as question-answering (QA) over long documents, but they tend to miss important information in the middle of context documents [(Liu 2023)](https://arxiv.org/abs/2307.03172). Here, we introduce *R&R*—a combination of two novel prompt-based methods called *reprompting* and *in-context retrieval* (ICR)—to alleviate this effect in document-based QA. In reprompting, we repeat the prompt instructions periodically throughout the context document to remind the LLM of its original task. In ICR, rather than instructing the LLM to answer the question directly, we instruct it to retrieve the top k passage numbers most relevant to the given question, which are then used as an abbreviated context in a second QA prompt. We test R&R with GPT-4 Turbo and Claude-2.1 on documents up to 80k tokens in length and observe a 16-point boost in QA accuracy on average. Our further analysis suggests that R&R improves performance on long document-based QA because it reduces the distance between relevant context and the instructions. Finally, we show that compared to short-context chunkwise methods, R&R enables the use of larger chunks that cost fewer LLM calls and output tokens, while minimizing the drop in accuracy.
Prompt-based learning is susceptible to intrinsic bias present in pre-trained language models (LMs), leading to sub-optimal performance in prompt-based zero/few-shot settings. In this work, we propose a null-input prompting method to calibrate intrinsic bias encoded in pre-trained LMs. Different from prior efforts that address intrinsic bias primarily for social fairness and often involve excessive computational cost, our objective is to explore enhancing LMs’ performance in downstream zero/few-shot learning while emphasizing the efficiency of intrinsic bias calibration. Specifically, we leverage a diverse set of auto-selected null-meaning inputs generated from GPT-4 to probe intrinsic bias of pre-trained LMs. Utilizing the bias-reflected probability distribution, we formulate a distribution disparity loss for bias calibration, where we exclusively update bias parameters (0.1% of total parameters) of LMs towards equal probability distribution. Experimental results show that the calibration promotes an equitable starting point for LMs while preserving language modeling abilities. Across a wide range of datasets, including sentiment analysis and topic classification, our method significantly improves zero/few-shot learning performance of LMs for both in-context learning and prompt-based fine-tuning (on average 9% and 2%, respectively).
Many open-ended conversations (e.g., tutoring lessons or business meetings) revolve around pre-defined reference materials, like worksheets or meeting bullets. To provide a framework for studying such conversation structure, we introduce *Problem-Oriented Segmentation & Retrieval (POSR), the task of jointly breaking down conversations into segments and linking each segment to the relevant reference item. As a case study, we apply POSR to education where effectively structuring lessons around problems is critical yet difficult. We present *LessonLink*, the first dataset of real-world tutoring lessons, featuring 3,500 segments, spanning 24,300 minutes of instruction and linked to 116 SAT Math problems. We define and evaluate several joint and independent approaches for POSR, including segmentation (e.g., TextTiling), retrieval (e.g., ColBERT), and large language models (LLMs) methods. Our results highlight that modeling POSR as one joint task is essential: POSR methods outperform independent segmentation and retrieval pipelines by up to +76% on joint metrics and surpass traditional segmentation methods by up to +78% on segmentation metrics. We demonstrate POSR’s practical impact on downstream education applications, deriving new insights on the language and time use in real-world lesson structures.
Large language models (LLMs) offer significant potential as tools to support an expanding range of decision-making tasks. Given their training on human (created) data, LLMs have been shown to inherit societal biases against protected groups, as well as be subject to bias functionally resembling cognitive bias. Human-like bias can impede fair and explainable decisions made with LLM assistance. Our work introduces BiasBuster, a framework designed to uncover, evaluate, and mitigate cognitive bias in LLMs, particularly in high-stakes decision-making tasks. Inspired by prior research in psychology and cognitive science, we develop a dataset containing 13,465 prompts to evaluate LLM decisions on different cognitive biases (e.g., prompt-induced, sequential, inherent). We test various bias mitigation strategies, while proposing a novel method utilizing LLMs to debias their own human-like cognitive bias within prompts. Our analysis provides a comprehensive picture of the presence and effects of cognitive bias across commercial and open-source models. We demonstrate that our selfhelp debiasing effectively mitigates model answers that display patterns akin to human cognitive bias without having to manually craft examples for each bias.
Pre-trained Language Models Return Distinguishable Probability Distributions to Unfaithfully Hallucinated Texts
PDF ↗In this work, we show the pre-trained language models return distinguishable generation probability and uncertainty distribution to unfaithfully hallucinated texts, regardless of their size and structure. By examining 24 models on 6 data sets, we find out that 88-98% of cases return statistically significantly distinguishable generation probability and uncertainty distributions. Using this general phenomenon, we showcase a hallucination-reducing training algorithm. Our algorithm outperforms other baselines by achieving higher faithfulness metrics while maintaining sound general text quality measures.
README: Bridging Medical Jargon and Lay Understanding for Patient Education through Data-Centric NLP
PDF ↗The advancement in healthcare has shifted focus toward patient-centric approaches, particularly in self-care and patient education, facilitated by access to Electronic Health Records (EHR). However, medical jargon in EHRs poses significant challenges in patient comprehension. To address this, we introduce a new task of automatically generating lay definitions, aiming to simplify complex medical terms into patient-friendly lay language. We first created the README dataset, an extensive collection of over 50,000 unique (medical term, lay definition) pairs and 300,000 mentions, each offering context-aware lay definitions manually annotated by domain experts. We have also engineered a data-centric Human-AI pipeline that synergizes data filtering, augmentation, and selection to improve data quality. We then used README as the training data for models and leveraged a Retrieval-Augmented Generation method to reduce hallucinations and improve the quality of model outputs. Our extensive automatic and human evaluations demonstrate that open-source mobile-friendly models, when fine-tuned with high-quality data, are capable of matching or even surpassing the performance of state-of-the-art closed-source large language models like ChatGPT. This research represents a significant stride in closing the knowledge gap in patient education and advancing patient-centric healthcare solutions.
The success of deep learning models on multi-hop fact verification has prompted researchers to understand the behavior behind their veracity. One feasible way is erasure search: obtaining the rationale by entirely removing a subset of input without compromising verification accuracy. Despite extensive exploration, current rationalization methods struggle to discern nuanced composition within the correlated evidence, which inevitably leads to noise rationalization in multi-hop scenarios. To address this issue, this paper explores the multi-granular rationale extraction method, aiming to realize the denoising rationalization for multi-hop fact verification. Specifically, given a pretrained veracity prediction model, two independent external explainers are introduced and trained collaboratively to enhance the discriminating ability by imposing varied constraints. Meanwhile, three key properties (Fidelity, Consistency, Salience) are introduced to regularize the denoising and faithful rationalization process. Additionally, a new Noiselessness metric is proposed to measure the purity of the rationales. Experimental results on three multi-hop fact verification datasets show that the proposed approach outperforms 12 baselines.
With recommender systems broadly deployed in various online platforms, many efforts have been devoted to learning user preferences and building effective sequential recommenders. However, existing work mainly focuses on capturing user implicit preferences from historical interactions and simply matching them with the next behavior, instead of predicting user explicit intentions. This may lead to inappropriate recommendations. In light of this issue, we propose the adversarial user intention learning approach for sequential recommendaiton, named AuriSRec. The major novelty of our approach is to explicitly predict user current intentions when making recommendations, by inferring their decision-making process as explained in target reviews (reviews written after interacting with the ground-truth item). Specifically, AuriSRec conducts adversarial learning between an intention generator and a discriminator. The generator predicts user intentions by taking their historical reviews and behavioral sequences as inputs, while target reviews provide guidance. Beyond typical sequential modeling methods in the field of natural language process (NLP), a decoupling-based review encoder and a hybrid attention fusion mechanism are introduced to filter noise and enhance the generation capacity. On the other hand, the discriminator determines whether the intention is generated or real based on their matching degree to the target item, thereby guiding the generator to produce gradually improved intentions. Extensive experiments on five real-world datasets demonstrate the effectiveness of our approach.
Dataset distillation is a process aimed at condensing datasets while preserving essential characteristics. In the text domain, prevailing methods typically generate distilled data as embedding vectors, which are not human-readable. This approach simplifies optimization but limits the transferability of distilled data across different model architectures. To address this limitation, we introduce a model-agnostic, data-efficient method that leverages Language Model (LM) embeddings. Compared to parameter-efficient methods such as LORA, our approach achieves comparable performance with significantly faster processing times. We evaluate our methodology through classification tasks on datasets like IMDB and AG-News, demonstrating performance that is on par with or exceeds previous model-dependent techniques. By utilizing LM embeddings, our method offers enhanced flexibility and improved transferability, expanding the range of potential applications.
Position bias in large language models (LLMs) leads to difficulty in accessing information retrieved from the retriever, thus downgrading the effectiveness of Retrieval-Augmented Generation (RAG) approaches in open-question answering. Recent studies reveal that this bias is related to disproportional attention across the context. In this work, we examine how to direct LLMs to allocate more attention towards a selected segment of the context through prompting, aiming to compensate for the shortage of attention. We find that language models do not have relative position awareness of the context but can be directed by promoting instruction with an exact document index. Our analysis contributes to a deeper understanding of position bias in LLMs and provides a pathway to mitigate this bias by instruction, thus benefiting LLMs in locating and utilizing relevant information from retrieved documents in RAG applications. The code and data in our study have been made publicly available.
Diverse and Effective Synthetic Data Generation for Adaptable Zero-Shot Dialogue State Tracking
PDF ↗We demonstrate substantial performance gains in zero-shot dialogue state tracking (DST) by enhancing training data diversity through synthetic data generation.Existing DST datasets are severely limited in the number of application domains and slot types they cover due to the high costs of data collection, restricting their adaptability to new domains.This work addresses this challenge with a novel, fully automatic data generation approach that creates synthetic zero-shot DST datasets.Distinguished from previous methods, our approach can generate dialogues across a massive range of application domains, complete with silver-standard dialogue state annotations and slot descriptions.This technique is used to create the D0T dataset for training zero-shot DST models, encompassing an unprecedented 1,000+ domains. Experiments on the MultiWOZ benchmark show that training models on diverse synthetic data improves Joint Goal Accuracy by 6.7%, achieving results competitive with models 13.5 times larger than ours.
“Knowing When You Don’t Know”: A Multilingual Relevance Assessment Dataset for Robust Retrieval-Augmented Generation
PDF ↗Retrieval-Augmented Generation (RAG) grounds Large Language Model (LLM) output by leveraging external knowledge sources to reduce factual hallucinations. However, prior work lacks a comprehensive evaluation of different language families, making it challenging to evaluate LLM robustness against errors in external retrieved knowledge. To overcome this, we establish **NoMIRACL**, a human-annotated dataset for evaluating LLM robustness in RAG across 18 typologically diverse languages. NoMIRACL includes both a non-relevant and a relevant subset. Queries in the non-relevant subset contain passages judged as non-relevant, whereas queries in the relevant subset include at least a single judged relevant passage. We measure relevance assessment using: (i) *hallucination rate*, measuring model tendency to hallucinate when the answer is not present in passages in the non-relevant subset, and (ii) *error rate*, measuring model inaccuracy to recognize relevant passages in the relevant subset. In our work, we observe that most models struggle to balance the two capacities. Models such as LLAMA-2 and Orca-2 achieve over 88% hallucination rate on the non-relevant subset. Mistral and LLAMA-3 hallucinate less but can achieve up to a 74.9% error rate on the relevant subset. Overall, GPT-4 is observed to provide the best tradeoff on both subsets, highlighting future work necessary to improve LLM robustness. NoMIRACL dataset and evaluation code are available at: https://github.com/project-miracl/nomiracl.
Enhancing Discourse Dependency Parsing with Sentence Dependency Parsing: A Unified Generative Method Based on Code Representation
PDF ↗Due to the high complexity of Discourse Dependency Parsing (DDP) tasks, their existing annotation resources are relatively scarce compared to other NLP tasks, and different DDP tasks also have significant differences in annotation schema. These issues have led to the dilemma of low resources for DDP tasks. Thanks to the powerful capabilities of Large Language Models (LLMs) in cross-task learning, we can use LLMs to model dependency parsing under different annotation schema in an unified manner, in order to alleviate the dilemma of low resources for DDP tasks. However, enabling LLMs to deeply comprehend dependency parsing tasks is a challenge that remains underexplored. Inspired by the application of code-based methods in complex tasks, we propose a code-based unified dependency parsing method. We treat the process of dependency parsing as a search process of dependency paths and use code to represent this search process. Furthermore, we use a curriculum-learning based instruction tuning strategy for joint training of multiple dependency parsing tasks. The experimental results show that our proposed code-based DDP system has achieved good performance on two Chinese DDP tasks (especially significant improvement on the DDP task with relatively less training data).
PizzaCommonSense: A Dataset for Commonsense Reasoning about Intermediate Steps in Cooking Recipes
PDF ↗Understanding procedural texts, such as cooking recipes, is essential for enabling machines to follow instructions and reason about tasks, a key aspect of intelligent reasoning. In cooking, these instructions can be interpreted as a series of modifications to a food preparation.For a model to effectively reason about cooking recipes, it must accurately discern and understand the inputs and outputs of intermediate steps within the recipe.We present a new corpus of cooking recipes enriched with descriptions of intermediate steps that describe the input and output for each step. PizzaCommonsense serves as a benchmark for the reasoning capabilities of LLMs because it demands rigorous explicit input-output descriptions to demonstrate the acquisition of implicit commonsense knowledge, which is unlikely to beeasily memorized. GPT-4 achieves only 26% human-evaluated preference for generations, leaving room for future improvements.
This study explores enhancing grammatical error correction (GEC) through automatic error generation (AEG) using language models (LMs). Specifically, we fine-tune Llama 2 LMs for error generation and find that this approach yields synthetic errors akin to human errors. Next, we train GEC Llama models using these artificial errors and outperform previous state-of-the-art error correction models, with gains ranging between 0.8 and 6 F0.5 points across all tested languages (German, Ukrainian, and Estonian). Moreover, we demonstrate that generating errors by fine-tuning smaller sequence-to-sequence models and prompting large commercial LMs (GPT3.5 and GPT4) also results in synthetic errors beneficially affecting error generation models. We openly release trained models for error generation and correction as well as all the synthesized error datasets for the covered languages.
Medical knowledge is context-dependent and requires consistent reasoning across various natural language expressions of semantically equivalent phrases. This is particularly crucial for drug names, where patients often use brand names like Advil or Tylenol instead of their generic equivalents. To study this, we create a new robustness dataset, RABBITS, to evaluate performance differences on medical benchmarks after swapping brand and generic drug names using physician expert annotations.We assess both open-source and API-based LLMs on MedQA and MedMCQA, revealing a consistent performance drop ranging from 1-10%. Furthermore, we identify a potential source of this fragility as the contamination of test data in widely used pre-training datasets.
Language Models for text classification often produce overconfident predictions for both in-distribution and out-of-distribution samples, i.e., the model’s output probabilities do not match their accuracy. Prior work showed that simple post-hoc approaches are effective for mitigating this issue, but are not robust in noisy settings, e.g., when the distribution shift is caused by spelling mistakes. In this work, we propose Distance Aware Calibration (DAC), a post-hoc approach that changes the confidence scores of a Language Model leveraging the distance between new samples been evaluated and the in-domain training set. We show that using DAC on top of a Language Model can improve in-domain calibration, robustness to different kind of distribution shift and also the model’s ability to detect out-of-distribution samples. We provide an extensive evaluation on common text classification benchmark for both calibration and out-of-distribution detection tasks.
The deployment of large language models (LLMs) is often constrained by memory bandwidth, where the primary bottleneck is the cost of transferring model parameters from the GPU’s global memory to its registers. When coupled with custom kernels that fuse the dequantization and matmul operations, weight-only quantization can thus enable faster inference by reducing the amount of memory movement. However, developing high-performance kernels for weight-quantized LLMs presents substantial challenges, especially when the weights are compressed to non-evenly-divisible bit widths (e.g., 3 bits) with non-uniform, lookup table (LUT) quantization. This paper describes FLUTE, a flexible lookup table engine for LUT-quantized LLMs, which uses offline restructuring of the quantized weight matrix to minimize bit manipulations associated with unpacking, and vectorization and duplication of the lookup table to mitigate shared memory bandwidth constraints. At batch sizes < 32 and quantization group size of 128 (typical in LLM inference), the FLUTE kernel can be 2-4x faster than existing GEMM kernels. As an application of FLUTE, we explore a simple extension to lookup table-based NormalFloat quantization and apply it to quantize LLaMA3 to various configurations, obtaining competitive quantization performance against strong baselines while obtaining an end-to-end throughput increase of 1.5 to 2 times.