Media outlets are becoming more partisan and polarized nowadays. In this paper, we identify media bias at the sentence level, and pinpoint bias sentences that intend to sway readers' opinions. As bias sentences are often expressed in a neutral and factual way, considering broader context outside a sentence can help reveal the bias. In particular, we observe that events in a bias sentence need to be understood in associations with other events in the document. Therefore, we propose to construct an event relation graph to explicitly reason about event-event relations for sentence-level bias identification. The designed event relation graph consists of events as nodes and four common types of event relations: coreference, temporal, causal, and subevent relations. Then, we incorporate event relation graph for bias sentences identification in two steps: an event-aware language model is built to inject the events and event relations knowledge into the basic language model via soft labels; further, a relation-aware graph attention network is designed to update sentence embedding with events and event relations information based on hard labels. Experiments on two benchmark datasets demonstrate that our approach with the aid of event relation graph improves both precision and recall of bias sentence identification.
Do transformers "think ahead" during inference at a given position? It is known transformers prepare information in the hidden states of the forward pass at time step $t$ that is then used in future forward passes $t+\tau$. We posit two explanations for this phenomenon: pre-caching, in which off-diagonal gradient terms present during training result in the model computing features at $t$ irrelevant to the present inference task but useful for the future, and breadcrumbs, in which features most relevant to time step $t$ are already the same as those that would most benefit inference at time $t+\tau$. We test these hypotheses by training language models without propagating gradients to past timesteps, a scheme we formalize as myopic training. In a constructed synthetic data setting, we find clear evidence for pre-caching. In the autoregressive language modeling setting, our experiments are more suggestive of the breadcrumbs hypothesis, though pre-caching increases with model scale.
Generative vision-language models (VLMs) have shown impressive performance in zero-shot vision-language tasks like image captioning and visual question answering. However, improving their zero-shot reasoning typically requires second-stage instruction tuning, which relies heavily on human-labeled or large language model-generated annotation, incurring high labeling costs. To tackle this challenge, we introduce Image-Conditioned Caption Correction (ICCC), a novel pre-training task designed to enhance VLMs' zero-shot performance without the need for labeled task-aware data. The ICCC task compels VLMs to rectify mismatches between visual and language concepts, thereby enhancing instruction following and text generation conditioned on visual inputs. Leveraging language structure and a lightweight dependency parser, we construct data samples of ICCC task from image-text datasets with low labeling and computation costs. Experimental results on BLIP-2 and InstructBLIP demonstrate significant improvements in zero-shot image-text generation-based VL tasks through ICCC instruction tuning.
Large-scale contrastive vision-language pre-trained models provide the zero-shot model achieving competitive performance across a range of image classification tasks without requiring training on downstream data. Recent works have confirmed that while additional fine-tuning of the zero-shot model on the reference data results in enhanced downstream performance, it compromises the model's robustness against distribution shifts. Our investigation begins by examining the conditions required to achieve the goals of robust fine-tuning, employing descriptions based on feature distortion theory and joint energy-based models. Subsequently, we propose a novel robust fine-tuning algorithm, Lipsum-FT, that effectively utilizes the language modeling aspect of the vision-language pre-trained models. Extensive experiments conducted on distribution shift scenarios in DomainNet and ImageNet confirm the superiority of our proposed Lipsum-FT approach over existing robust fine-tuning methods.
Medical image analysis is essential to clinical diagnosis and treatment, which is increasingly supported by multi-modal large language models (MLLMs). However, previous research has primarily focused on 2D medical images, leaving 3D images under-explored, despite their richer spatial information. This paper aims to advance 3D medical image analysis with MLLMs. To this end, we present a large-scale 3D multi-modal medical dataset, M3D-Data, comprising 120K image-text pairs and 662K instruction-response pairs specifically tailored for various 3D medical tasks, such as image-text retrieval, report generation, visual question answering, positioning, and segmentation. Additionally, we propose M3D-LaMed, a versatile multi-modal large language model for 3D medical image analysis. Furthermore, we introduce a new 3D multi-modal medical benchmark, M3D-Bench, which facilitates automatic evaluation across eight tasks. Through comprehensive evaluation, our method proves to be a robust model for 3D medical image analysis, outperforming existing solutions. All code, data, and models are publicly available at: https://github.com/BAAI-DCAI/M3D.
Text-to-image (T2I) generative models have attracted significant attention and found extensive applications within and beyond academic research. For example, the Civitai community, a platform for T2I innovation, currently hosts an impressive array of 74,492 distinct models. However, this diversity presents a formidable challenge in selecting the most appropriate model and parameters, a process that typically requires numerous trials. Drawing inspiration from the tool usage research of large language models (LLMs), we introduce DiffAgent, an LLM agent designed to screen the accurate selection in seconds via API calls. DiffAgent leverages a novel two-stage training framework, SFTA, enabling it to accurately align T2I API responses with user input in accordance with human preferences. To train and evaluate DiffAgent's capabilities, we present DABench, a comprehensive dataset encompassing an extensive range of T2I APIs from the community. Our evaluations reveal that DiffAgent not only excels in identifying the appropriate T2I API but also underscores the effectiveness of the SFTA training framework. Codes are available at https://github.com/OpenGVLab/DiffAgent.
Contemporary conversational systems often present a significant limitation: their responses lack the emotional depth and disfluent characteristic of human interactions. This absence becomes particularly noticeable when users seek more personalized and empathetic interactions. Consequently, this makes them seem mechanical and less relatable to human users. Recognizing this gap, we embarked on a journey to humanize machine communication, to ensure AI systems not only comprehend but also resonate. To address this shortcoming, we have designed an innovative speech synthesis pipeline. Within this framework, a cutting-edge language model introduces both human-like emotion and disfluencies in a zero-shot setting. These intricacies are seamlessly integrated into the generated text by the language model during text generation, allowing the system to mirror human speech patterns better, promoting more intuitive and natural user interactions. These generated elements are then adeptly transformed into corresponding speech patterns and emotive sounds using a rule-based approach during the text-to-speech phase. Based on our experiments, our novel system produces synthesized speech that's almost indistinguishable from genuine human communication, making each interaction feel more personal and authentic.
This paper introduces Evalverse, a novel library that streamlines the evaluation of Large Language Models (LLMs) by unifying disparate evaluation tools into a single, user-friendly framework. Evalverse enables individuals with limited knowledge of artificial intelligence to easily request LLM evaluations and receive detailed reports, facilitated by an integration with communication platforms like Slack. Thus, Evalverse serves as a powerful tool for the comprehensive assessment of LLMs, offering both researchers and practitioners a centralized and easily accessible evaluation framework. Finally, we also provide a demo video for Evalverse, showcasing its capabilities and implementation in a two-minute format.
The recent advancements in large language models (LLMs) have revolutionized the field of natural language processing, progressively broadening their scope to multimodal perception and generation. However, effectively integrating listening capabilities into LLMs poses significant challenges, particularly with respect to generalizing across varied contexts and executing complex auditory tasks. In this work, we introduce WavLLM, a robust and adaptive speech large language model with dual encoders, and a prompt-aware LoRA weight adapter, optimized by a two-stage curriculum learning approach. Leveraging dual encoders, we decouple different types of speech information, utilizing a Whisper encoder to process the semantic content of speech, and a WavLM encoder to capture the unique characteristics of the speaker's identity. Within the curriculum learning framework, WavLLM first builds its foundational capabilities by optimizing on mixed elementary single tasks, followed by advanced multi-task training on more complex tasks such as combinations of the elementary tasks. To enhance the flexibility and adherence to different tasks and instructions, a prompt-aware LoRA weight adapter is introduced in the second advanced multi-task training stage. We validate the proposed model on universal speech benchmarks including tasks such as ASR, ST, SV, ER, and also apply it to specialized datasets like Gaokao English listening comprehension set for SQA, and speech Chain-of-Thought (CoT) evaluation set. Experiments demonstrate that the proposed model achieves state-of-the-art performance across a range of speech tasks on the same model size, exhibiting robust generalization capabilities in executing complex tasks using CoT approach. Furthermore, our model successfully completes Gaokao tasks without specialized training. The codes, models, audio, and Gaokao evaluation set can be accessed at \url{aka.ms/wavllm}.
By training to predict the next token in an unlabeled corpus, large language models learn to perform many tasks without any labeled data. However, their next-token-prediction objective arguably limits their performance in scenarios that require planning, such as writing a coherent article. In this paper, we train a module for planning the future writing process via a self-supervised learning objective. Given the textual context, this planning module learns to predict future abstract writing actions, which correspond to centroids in a clustered text embedding space. By conditioning on these actions, our model extends the successful language model formula to more abstract planning in an unsupervised way. Empirically, we demonstrate that our method improves language modeling performance in general, particularly with respect to the text structure. Because our framework uses a planner module that is unsupervised and external to the language model, new planner modules can be trained at large scale and easily be shared with the community.
Reinforcement learning from human feedback (RLHF) has been a central technique for recent large language model (LLM) alignment. However, its heavy dependence on costly human or LLM-as-Judge preference feedback could stymie its wider applications. In this work, we introduce Self-Contrast, a feedback-free large language model alignment method via exploiting extensive self-generated negatives. With only supervised fine-tuning (SFT) targets, Self-Contrast leverages the LLM itself to generate massive diverse candidates, and harnesses a pre-trained embedding model to filter multiple negatives according to text similarity. Theoretically, we illustrate that in this setting, merely scaling negative responses can still effectively approximate situations with more balanced positive and negative preference annotations. Our experiments with direct preference optimization (DPO) on three datasets show that, Self-Contrast could consistently outperform SFT and standard DPO training by large margins. And as the number of self-generated negatives increases, the performance of Self-Contrast continues to grow. Code and data are available at https://github.com/THUDM/Self-Contrast.
State-of-the-art language model fine-tuning techniques, such as Direct Preference Optimization (DPO), restrict user control by hard-coding predefined behaviors into the model. To address this, we propose a novel method, Configurable Safety Tuning (CST), that augments DPO using synthetic preference data to facilitate flexible safety configuration of LLMs at inference time. CST overcomes the constraints of vanilla DPO by introducing a system prompt specifying safety configurations, enabling LLM deployers to disable/enable safety preferences based on their need, just changing the system prompt. Our experimental evaluations indicate that CST successfully manages different safety configurations and retains the original functionality of LLMs, showing it is a robust method for configurable deployment. Data and models available at https://github.com/vicgalle/configurable-safety-tuning
Off-the-shelf pre-trained language models have become the de facto standard in NLP pipelines for a multitude of downstream tasks. However, the inability of these models to properly encode numerals limits their performance on tasks requiring numeric comprehension. We introduce strategies to semantically prime numerals in any corpus by generating anchors governed by the distribution of numerals in said corpus, thereby enabling mathematically grounded representations of these numeral tokens. We establish the superiority of our proposed techniques through evaluation on a range of numeracy tasks for both in-domain (seen) and out-domain (unseen) numerals. Further, we expand our empirical evaluations to numerals ranging from 1 to 10 billion, a significantly broader range compared to previous studies of the same nature, and we demonstrate significant improvements in the mathematical grounding of our learned embeddings.
Preference modeling techniques, such as direct preference optimization (DPO), has shown effective in enhancing the generalization abilities of large language model (LLM). However, in tasks involving video instruction-following, providing informative feedback, especially for detecting hallucinations in generated responses, remains a significant challenge. Previous studies have explored using large large multimodal models (LMMs) as reward models to guide preference modeling, but their ability to accurately assess the factuality of generated responses compared to corresponding videos has not been conclusively established. This paper introduces a novel framework that utilizes detailed video captions as a proxy of video content, enabling language models to incorporate this information as supporting evidence for scoring video Question Answering (QA) predictions. Our approach demonstrates robust alignment with OpenAI GPT-4V model's reward mechanism, which directly takes video frames as input. Furthermore, we show that applying this tailored reward through DPO significantly improves the performance of video LMMs on video QA tasks.
Instruction tuning is an important step in making language models useful for direct user interaction. However, the legal domain is underrepresented in typical instruction datasets (e.g., only 10 out of 1600+ tasks in Super-NaturalInstructions). To study whether instruction tuning on legal datasets is necessary for strong legal reasoning, we aggregate 58 annotated legal datasets and write instructions for each, creating LawInstruct. LawInstruct covers 17 global jurisdictions, 24 languages and a total of 12M examples across diverse tasks such as legal QA, summarization of court cases, and legal argument mining. We evaluate our models on LegalBench, measuring legal reasoning across five categories in 162 challenging and realistic legal tasks, and MMLU, to measure potential drops in general reasoning capabilities. We find that legal-specific instruction tuning on Flan-T5 - yielding FLawN-T5 - improves performance on LegalBench across all model sizes, with an aggregate increase of 15 points or 50% over Flan-T5 for the base size. No model size shows performance drops in MMLU. We publish LawInstruct as a resource for further study of instruction tuning in the legal domain.
The development of game agents holds a critical role in advancing towards Artificial General Intelligence. The progress of Large Language Models (LLMs) offers an unprecedented opportunity to evolve and empower game agents with human-like decision-making capabilities in complex computer game environments. This paper provides a comprehensive overview of LLM-based game agents from a holistic viewpoint. First, we introduce the conceptual architecture of LLM-based game agents, centered around three core functional components: memory, reasoning and in/output. Second, we survey existing representative LLM-based game agents documented in the literature with respect to methodologies and adaptation agility across six genres of games, including adventure, communication, competition, cooperation, simulation, and crafting & exploration games. Finally, we present an outlook of future research and development directions in this burgeoning field. A curated list of relevant papers is maintained and made accessible at: https://github.com/git-disl/awesome-LLM-game-agent-papers.
Although there has been rapid progress in endowing robots with the ability to solve complex manipulation tasks, generating control policies for bimanual robots to solve tasks involving two hands is still challenging because of the difficulties in effective temporal and spatial coordination. With emergent abilities in terms of step-by-step reasoning and in-context learning, Large Language Models (LLMs) have demonstrated promising potential in a variety of robotic tasks. However, the nature of language communication via a single sequence of discrete symbols makes LLM-based coordination in continuous space a particular challenge for bimanual tasks. To tackle this challenge, we present LAnguage-model-based Bimanual ORchestration (LABOR), an agent utilizing an LLM to analyze task configurations and devise coordination control policies for addressing long-horizon bimanual tasks. We evaluate our method through simulated experiments involving two classes of long-horizon tasks using the NICOL humanoid robot. Our results demonstrate that our method outperforms the baseline in terms of success rate. Additionally, we thoroughly analyze failure cases, offering insights into LLM-based approaches in bimanual robotic control and revealing future research trends. The project website can be found at http://labor-agent.github.io.
Best-of-N (BoN) sampling with a reward model has been shown to be an effective strategy for aligning Large Language Models (LLMs) to human preferences at the time of decoding. BoN sampling is susceptible to a problem known as reward hacking when the accuracy of the reward model is not high enough due to the quality or the quantity of the preference dataset. Because the reward model is an imperfect proxy for the true objective, over-optimizing its value can compromise its performance on the true objective. In this research, we propose MBR-BoN, a variant of BoN that aims to mitigate reward hacking at inference time by incorporating the Minimum Bayes Risk (MBR) objective as a proximity regularization term. We show empirically and analytically that the MBR objective quantifies the proximity of the response to the reference policy, serving as a proximity regularizer. We evaluate MBR-BoN on the AlpacaFarm and Anthropic's hh-rlhf datasets and show that it outperforms both BoN sampling and MBR decoding. We also evaluate MBR-BoN to generate a pairwise preference learning dataset for Direct Preference Optimization (DPO). Empirical results show that models trained on a dataset generated with MBR-BoN outperform those with vanilla BoN. Our code is available at https://github.com/CyberAgentAILab/regularized-bon
We explored leveraging state-of-the-art deep learning, big data, and natural language processing to enhance the detection of vulnerable web server versions. Focusing on improving accuracy and specificity over rule-based systems, we conducted experiments by sending various ambiguous and non-standard HTTP requests to 4.77 million domains and capturing HTTP response status lines. We represented these status lines through training a BPE tokenizer and RoBERTa encoder for unsupervised masked language modeling. We then dimensionality reduced and concatenated encoded response lines to represent each domain's web server. A Random Forest and multilayer perceptron (MLP) classified these web servers, and achieved 0.94 and 0.96 macro F1-score, respectively, on detecting the five most popular origin web servers. The MLP achieved a weighted F1-score of 0.55 on classifying 347 major type and minor version pairs. Analysis indicates that our test cases are meaningful discriminants of web server types. Our approach demonstrates promise as a powerful and flexible alternative to rule-based systems.
System-prompting is a standard tool for customizing language-model chatbots, enabling them to follow a specific instruction. An implicit assumption in the use of system prompts is that they will be stable, so the chatbot will continue to generate text according to the stipulated instructions for the duration of a conversation. We propose a quantitative benchmark to test this assumption, evaluating instruction stability via self-chats between two instructed chatbots. Testing popular models like LLaMA2-chat-70B and GPT-3.5, we reveal a significant instruction drift within eight rounds of conversations. An empirical and theoretical analysis of this phenomenon suggests the transformer attention mechanism plays a role, due to attention decay over long exchanges. To combat attention decay and instruction drift, we propose a lightweight method called split-softmax, which compares favorably against two strong baselines.