United States

In this work, we designed unbiased prompts to systematically evaluate the psychological safety of large language models (LLMs). First, we tested five different LLMs by using two personality tests: Short Dark Triad (SD-3) and Big Five Inventory (BFI). All models scored higher than the human average on SD-3, suggesting a relatively darker personality pattern. Despite being instruction fine-tuned with safety metrics to reduce toxicity, InstructGPT, GPT-3.5, and GPT-4 still showed dark personality patterns; these models scored higher than self-supervised GPT-3 on the Machiavellianism and narcissism traits on SD-3. Then, we evaluated the LLMs in the GPT series by using well-being tests to study the impact of fine-tuning with more training data. We observed a continuous increase in the well-being scores of GPT models. Following these observations, we showed that fine-tuning Llama-2-chat-7B with responses from BFI using direct preference optimization could effectively reduce the psychological toxicity of the model. Based on the findings, we recommended the application of systematic and comprehensive psychological metrics to further evaluate and improve the safety of LLMs.

EMNLP 2024

Evaluating Psychological Safety of Large Language Models

low resource

data augmentation

sentiment classification

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

poster

Unsupervised multitask pre-training has been the critical method behind the recent success of language models (LMs). However, supervised multitask learning still holds significant promise, as scaling it in the post-training stage trends towards better generalization. In this paper, we explore supervised multitask pre-training by proposing Instruction Pre-training, a framework that scalably augments massive raw corpora with instruction-response pairs to pre-train LMs. The instruction-response pairs are generated by an efficient instruction synthesizer built on open-source models. In our experiments, we synthesize 200M instruction response pairs covering 40+ task categories to verify the effectiveness of Instruction Pre-training. In pre-training from scratch, Instruction Pre-training not only consistently enhances pre-trained base models but also benefits more from further instruction tuning. In continual pre-training, Instruction Pre-training enables Llama3-8B to be comparable to or even outperform Llama3-70B. Our model, code, and data are available at https://github.com/microsoft/LMOps.

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the \textit{Uncertainty} about the question and the \textit{Fidelity} to the answer generated by language models. Then, we propose a plug-and-play method, \textit{UF Calibration}, to estimate the confidence of language models. Our method has shown good calibration performance by conducting experiments with 6 RLHF-LMs on four MCQA datasets. Moreover, we propose two novel metrics, IPR and CE, to evaluate the calibration of the model, and we have conducted a detailed discussion on \textit{Truly Well-Calibrated Confidence} for large language models. Our method could serve as a strong baseline, and we hope that this work will provide some insights into the model confidence calibration.

Calibrating the Confidence of Large Language Models by Eliciting Fidelity

Adversarial textual examples reveal the vulnerability of natural language processing (NLP) models. Most existing text attack methods are designed for English text, while the robust implementation of the second popular language, i.e., Chinese with 1 billion users, is greatly underestimated. Although several Chinese attack methods have been presented, they either directly transfer from English attacks or adopt simple greedy search to optimize the attack priority, usually leading to unnatural sentences. To address these issues, we propose an adaptive Immune-based Sound-Shape Code (ISSC) algorithm for adversarial Chinese text attacks. Firstly, we leverage the Sound-Shape code to generate natural substitutions, which comprehensively integrate multiple Chinese features. Secondly, we employ adaptive immune algorithm (IA) to determine the replacement order, which can reduce the duplication of population to improve the search ability. Extensive experimental results validate the superiority of our ISSC in producing high-quality Chinese adversarial texts. Our code and data can be found in https://github.com/nohuma/chinese-attack-issc.

Adaptive Immune-based Sound-Shape Code Substitution for Adversarial Chinese Text Attacks

The tool-use Large Language Models (LLMs) that integrate with external Python interpreters have significantly enhanced mathematical reasoning capabilities for open-source LLMs, while tool-free methods chose another track: augmenting math reasoning data. However, a great method to integrate the above two research paths and combine their advantages remains to be explored. In this work, we firstly include new math questions via **mu**lti-perspective data augmenting methods and then synthesize **code**-nested solutions to them. The open LLMs (e.g., Llama-2) are finetuned on the augmented dataset to get the resulting models, **MuMath-Code** ($\mu$-Math-Code). During the inference phase, our MuMath-Code generates code and interacts with the external python interpreter to get the execution results. Therefore, MuMath-Code leverages the advantages of both the external tool and data augmentation. To fully leverage the advantages of our augmented data, we propose a two-stage training strategy: In Stage-1, we finetune Llama-2 on pure CoT data to get an intermediate model, which then is trained on the code-nested data in Stage-2 to get the resulting MuMath-Code.
Our MuMath-Code-7B achieves 83.8\% on GSM8K and 52.4\% on MATH, while MuMath-Code-70B model achieves new state-of-the-art performance among open methods---achieving 90.7\% on GSM8K and 55.1\% on MATH. Extensive experiments validate the combination of tool use and data augmentation, as well as our two-stage training strategy.
We release the proposed dataset along with the associated code for public use: https://github.com/youweihao-tal/MuMath-Code.

MuMath-Code: Combining Tool-Use Large Language Models with Multi-perspective Data Augmentation for Mathematical Reasoning

Temporal Knowledge Graph (TKG) reasoning, aiming to predict future unknown facts based on historical information, has attracted considerable attention due to its great practical value. Insight into history is the key to predict the future. However, most existing TKG reasoning models singly capture repetitive history, ignoring the entity's multi-hop neighbour history which can provide valuable background knowledge for TKG reasoning. In this paper, we propose $\textbf{M}$ulti-$\textbf{G}$ranularity History and $\textbf{E}$ntity $\textbf{S}$imilarity $\textbf{L}$earning (MGESL) model for Temporal Knowledge Graph Reasoning, which models historical information from both coarse-grained and fine-grained history. Since similar entities tend to exhibit similar behavioural patterns, we also design a hypergraph convolution aggregator to capture the similarity between entities. Furthermore, we introduce a more realistic setting for the TKG reasoning, where candidate entities are already known at the timestamp to be predicted. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our proposed model.

Multi-Granularity History and Entity Similarity Learning for Temporal Knowledge Graph Reasoning

Large language models (LLMs) have shown excellent capability for solving reasoning problems. Existing approaches do not differentiate the question difficulty when designing prompting methods for them. Clearly, a simple method cannot elicit sufficient knowledge from LLMs to answer a hard question. Meanwhile, a sophisticated one will force the LLM to generate redundant or even inaccurate intermediate steps toward a simple question. Consequently, the performance of existing methods fluctuates among various questions.

In this work, we propose Adaption-of-Thought (AdoT), an adaptive method to improve LLMs for the reasoning problem, which first measures the question difficulty and then tailors demonstration set construction and difficulty-adapted retrieval strategies for the adaptive demonstration construction. Experimental results on three reasoning tasks prove the superiority of our proposed method, showing an absolute improvement of up to 5.5% on arithmetic reasoning, 7.4% on symbolic reasoning, and 2.3% on commonsense reasoning. Our codes and implementation details are available at: https://github.com/NLPGM/AdoT

Adaption-of-Thought: Learning Question Difficulty Improves Large Language Models for Reasoning

Transformer-based large language models (LLMs) exhibit limitations such as generating unsafe responses, unreliable reasoning, etc. Existing inference intervention approaches attempt to mitigate these issues by finetuning additional models to produce calibration signals (such as rewards) that guide the LLM's decoding process. However, this solution introduces substantial time and space overhead due to the separate models required. This work proposes Non-disruptive parameters insertion (Otter), inserting extra parameters into the transformer architecture to predict calibration signals along with the original LLM output. Otter offers state-of-the-art performance on multiple demanding tasks while saving up to 86.5% extra space and 98.5% extra time. Furthermore, Otter seamlessly integrates with existing inference engines, requiring only a one-line code change, and the original model response remains accessible after the parameter insertion.

Predicting Rewards Alongside Tokens: Non-disruptive Parameter Insertion for Efficient Inference Intervention in Large Language Model

Projecting visual features into word embedding space has become a significant fusion strategy adopted by Multimodal Large Language Models (MLLMs). However, its internal mechanisms have yet to be explored. Inspired by multilingual research, we identify domain-specific neurons in multimodal large language models. Specifically, we investigate the distribution of domain-specific neurons and the mechanism of how MLLMs process features from diverse domains. Furthermore, we propose a three-stage framework for language model modules in MLLMs when handling projected image features, and verify this hypothesis using logit lens. Extensive experiments indicate that while current MLLMs exhibit Visual Question Answering (VQA) capability, they may not fully utilize domain-specific information. Manipulating domain-specific neurons properly will result in a 10% change of accuracy at most, shedding light on the development of cross-domain, all-encompassing MLLMs in the future. The source code is available at https://anonymous.4open.science/r/MMNeuron.

MMNeuron: Discovering Neuron-Level Domain-Specific Interpretation in Multimodal Large Language Model

Vietnamese, a low-resource language, is typically categorized into three primary dialect groups that belong to Northern, Central, and Southern Vietnam. However, each province within these regions exhibits its own distinct pronunciation variations. Despite the existence of various speech recognition datasets, none of them has provided a fine-grained classification of the 63 dialects specific to individual provinces of Vietnam. To address this gap, we introduce Vietnamese Multi-Dialect (ViMD) dataset, a novel comprehensive dataset capturing the rich diversity of 63 provincial dialects spoken across Vietnam. Our dataset comprises 102.56 hours of audio, consisting of approximately 19,000 utterances, and the associated transcripts contain over 1.2 million words. To provide benchmarks and simultaneously demonstrate the challenges of our dataset, we fine-tune state-of-the-art pre-trained models for two downstream tasks: (1) Dialect identification and (2) Speech recognition. The empirical results suggest two implications including the influence of geographical factors on dialects, and the constraints of current approaches in speech recognition tasks involving multi-dialect speech data. Our dataset is available for research purposes.

Multi-Dialect Vietnamese: Task, Dataset, Baseline Models and Challenges

Reinforcement learning from human feedback (RLHF) and AI-generated feedback (RLAIF) have become prominent techniques that significantly enhance the functionality of pre-trained language models (LMs). These methods harness feedback, sourced either from humans or AI, as direct rewards or to shape reward models that steer LM optimization. Nonetheless, the effective integration of rewards from diverse sources presents a significant challenge due to their disparate characteristics. To address this, recent research has developed algorithms incorporating strategies such as weighting, ranking, and constraining to handle this complexity. Despite these innovations, a bias toward disproportionately high rewards can still skew the reinforcement learning process and negatively impact LM performance. This paper explores a methodology for reward composition that enables simultaneous improvements in LMs across multiple dimensions. Inspired by fairness theory, we introduce a training algorithm that aims to reduce disparity and enhance stability among various rewards. Our method treats the aggregate reward as a dynamic weighted sum of individual rewards, with alternating updates to the weights and model parameters. For efficient and straightforward implementation, we employ an estimation technique rooted in the mirror descent method for weight updates, eliminating the need for gradient computations. The empirical results under various types of rewards across a wide range of scenarios demonstrate the effectiveness of our method.

Downloads

Next from EMNLP 2024

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Similar lecture

On the Reliability of Psychological Scales on Large Language Models