Large Language Models (LLMs) have seen widespread adoption due to their remarkable natural language capabilities. However, when deploying them in real-world settings, it is important to align LLMs to generate texts according to acceptable human standards. Methods such as Proximal Policy Optimization (PPO) and Direct Preference Optimization (DPO) have made significant progress in refining LLMs using human preference data. However, the privacy concerns inherent in utilizing such preference data have yet to be adequately studied. In this paper, we investigate the vulnerability of LLMs aligned using human preference datasets to membership inference attacks (MIAs), highlighting the shortcomings of previous MIA approaches with respect to preference data. Our study has two main contributions: first, we introduce a novel reference-based attack framework specifically for analyzing preference data called PREMIA (\uline{Pre}ference data \uline{MIA}); second, we provide empirical evidence that DPO models are more vulnerable to MIA compared to PPO models. Our findings highlight gaps in current privacy-preserving practices for LLM alignment.

Enabling mobile robots to follow complex natural language instructions is an important yet challenging problem. People want to flexibly express constraints, refer to arbitrary landmarks and verify behavior when instructing robots. Conversely, robots must disambiguate human instructions into specifications and ground instruction referents in the real world. We propose Language Instruction grounding for Motion Planning (LIMP), an approach that enables robots to verifiably follow expressive and complex open-ended instructions in real-world environments without prebuilt semantic maps. LIMP constructs a symbolic instruction representation that reveals the robot's alignment with an instructor's intended motives and affords the synthesis of robot behaviors that are correct-by-construction. We perform a large scale evaluation and demonstrate our approach on 150 instructions in five real-world environments showing the generality of our approach and the ease of deployment in novel unstructured domains. In our experiments, LIMP performs comparably with state-of-the-art LLM task planners and LLM code-writing planners on standard open vocabulary tasks and additionally achieves 79\% success rate on complex spatiotemporal instructions while LLM and Code-writing planners both achieve 38\%. See supplementary materials and demo videos at https://robotlimp.github.io

Hierarchical control for robotics has long been plagued by the need to have a well defined interface layer to communicate between high-level task planners and low-level policies. With the advent of LLMs, language has been emerging as a prospective interface layer. However, this has several limitations. Not all tasks can be decomposed into steps that are easily expressible in natural language (e.g. performing a dance routine). Further, it makes end-to-end finetuning on embodied data challenging due to domain shift and catastrophic forgetting. We introduce our method -- Learnable Latent Codes as Bridges (LCB) -- as an alternate architecture to overcome these limitations. \method~uses a learnable latent code to act as a bridge between LLMs and low-level policies. This enables LLMs to flexibly communicate goals in the task plan without being entirely constrained by language limitations. Additionally, it enables end-to-end finetuning without destroying the embedding space of word tokens learned during pre-training. Through experiments on Language Table and Calvin, two common language based benchmarks for embodied agents, we find that \method~outperforms baselines (including those w/ GPT-4V) that leverage pure language as the interface layer on tasks that require reasoning and multi-step behaviors.

This work explores the capability of conversational chatbots powered by large language models (LLMs), to understand and characterize predicate symmetry, a cognitive linguistic function traditionally believed to be an inherent human trait. Leveraging in-context learning (ICL), a paradigm shift enabling chatbots to learn new tasks from prompts without re-training, we assess the symmetrical reasoning of five chatbots: ChatGPT 4, Huggingface chat AI, Microsoft's Copilot AI, LLaMA through Perplexity, and Gemini Advanced. Using the Symmetry Inference Sentence (SIS) dataset by Tanchip et al. (2020), we compare chatbot responses against human evaluations to gauge their understanding of predicate symmetry. Experiment results reveal varied performance among chatbots, with some approaching human-like reasoning capabilities. Gemini, for example, reaches a correlation of 0.85 with human scores, while providing a sounding justification for each symmetry evaluation. This study underscores the potential and limitations of LLMs in mirroring complex cognitive processes as symmetrical reasoning.

Large Language Models~(LLMs) demonstrate remarkable translation capabilities in high-resource language tasks, yet their performance in low-resource languages is hindered by insufficient multilingual data during pre-training. To address this, we dedicate 35,000 A100-SXM4-80GB GPU hours in conducting extensive multilingual continual pre-training on the LLaMA series models, enabling translation support across more than 100 languages. Through a comprehensive analysis of training strategies, such as vocabulary expansion and data augmentation, we develop LLaMAX. Remarkably, without sacrificing its generalization ability, LLaMAX achieves significantly higher translation performance compared to existing open-source LLMs~(by more than 10 spBLEU points) and performs on-par with specialized translation model~(M2M-100-12B) on the Flores-101 benchmark. Extensive experiments indicate that LLaMAX can serve as a robust multilingual foundation model. The code~\footnote{\url{https://github.com/CONE-MT/LLaMAX/.}} and models~\footnote{\url{https://huggingface.co/LLaMAX/.}} are publicly available.

LLM-based agents have demonstrated impressive zero-shot performance in the vision-language navigation (VLN) task. However, these zero-shot methods focus only on solving high-level task planning by selecting nodes in predefined navigation graphs for movements, overlooking low-level control in realistic navigation scenarios. To bridge this gap, we propose AO-Planner, a novel affordances-oriented planning framework for continuous VLN task. Our AO-Planner integrates various foundation models to achieve affordances-oriented motion planning and action decision-making, both performed in a zero-shot manner. Specifically, we employ a visual affordances prompting (VAP) approach, where visible ground is segmented utilizing SAM to provide navigational affordances, based on which the LLM selects potential next waypoints and generates low-level path planning towards selected waypoints. We further introduce a high-level agent, PathAgent, to identify the most probable pixel-based path and convert it into 3D coordinates to fulfill low-level motion. Experimental results on the challenging R2R-CE benchmark demonstrate that AO-Planner achieves state-of-the-art zero-shot performance (5.5% improvement in SPL). Our method establishes an effective connection between LLM and 3D world to circumvent the difficulty of directly predicting world coordinates, presenting novel prospects for employing foundation models in low-level motion control.

Harnessing the potential of large language models (LLMs) like ChatGPT can help address social challenges through inclusive, ethical, and sustainable means. In this paper, we investigate the extent to which ChatGPT can annotate data for social computing tasks, aiming to reduce the complexity and cost of undertaking web research. To evaluate ChatGPT's potential, we re-annotate seven datasets using ChatGPT, covering topics related to pressing social issues like COVID-19 misinformation, social bot deception, cyberbully, clickbait news, and the Russo-Ukrainian War. Our findings demonstrate that ChatGPT exhibits promise in handling these data annotation tasks, albeit with some challenges. Across the seven datasets, ChatGPT achieves an average annotation F1-score of 72.00%. Its performance excels in clickbait news annotation, correctly labeling 89.66% of the data. However, we also observe significant variations in performance across individual labels. Our study reveals predictable patterns in ChatGPT's annotation performance. Thus, we propose GPT-Rater, a tool to predict if ChatGPT can correctly label data for a given annotation task. Researchers can use this to identify where ChatGPT might be suitable for their annotation requirements. We show that GPT-Rater effectively predicts ChatGPT's performance. It performs best on a clickbait headlines dataset by achieving an average F1-score of 95.00%. We believe that this research opens new avenues for analysis and can reduce barriers to engaging in social computing research.

Mathematical verfier achieves success in mathematical reasoning tasks by validating the correctness of solutions. However, existing verifiers are trained with binary classification labels, which are not informative enough for the model to accurately assess the solutions. To mitigate the aforementioned insufficiency of binary labels, we introduce step-wise natural language feedbacks as rationale labels (i.e., the correctness of the current step and the explanations). In this paper, we propose \textbf{Math-Minos}, a natural language feedback enhanced verifier by constructing automatically-generated training data and a two-stage training paradigm for effective training and efficient inference. Our experiments reveal that a small set (30k) of natural language feedbacks can significantly boost the performance of the verifier by the accuracy of 1.6\% (86.6\% $\rightarrow$ 88.2\%) on GSM8K and 0.8\% (37.8\% $\rightarrow$ 38.6\%) on MATH. We have released our code and data for further exploration.

Large Language Models have found application in various mundane and repetitive tasks including Human Resource (HR) support. We worked with the domain experts of SAP SE to develop an HR support chatbot as an efficient and effective tool for addressing employee inquiries. We inserted a human-in-the-loop in various parts of the development cycles such as dataset collection, prompt optimization, and evaluation of generated output. By enhancing the LLM-driven chatbot's response quality and exploring alternative retrieval methods, we have created an efficient, scalable, and flexible tool for HR professionals to address employee inquiries effectively. Our experiments and evaluation conclude that GPT-4 outperforms other models and can overcome inconsistencies in data through internal reasoning capabilities. Additionally, through expert analysis, we infer that reference-free evaluation metrics such as G-Eval and Prometheus demonstrate reliability closely aligned with that of human evaluation.

Direct Preference Optimization (DPO) is a widely used offline preference optimization algorithm that reparameterizes reward functions in reinforcement learning from human feedback (RLHF) to enhance simplicity and training stability. In this work, we propose SimPO, a simpler yet more effective approach. The effectiveness of SimPO is attributed to a key design: using the average log probability of a sequence as the implicit reward. This reward formulation better aligns with model generation and eliminates the need for a reference model, making it more compute and memory efficient. Additionally, we introduce a target reward margin to the Bradley-Terry objective to encourage a larger margin between the winning and losing responses, further enhancing the algorithm's performance. We compare SimPO to DPO and its latest variants across various state-of-the-art training setups, including both base and instruction-tuned models like Mistral and Llama3. We evaluated on extensive instruction-following benchmarks, including AlpacaEval 2, MT-Bench, and the recent challenging Arena-Hard benchmark. Our results demonstrate that SimPO consistently and significantly outperforms existing approaches without substantially increasing response length. Specifically, SimPO outperforms DPO by up to 6.4 points on AlpacaEval 2 and by up to 7.5 points on Arena-Hard. Our top-performing model, built on Llama3-8B-Instruct, achieves a remarkable 53.7 length-controlled win rate on AlpacaEval 2 -- surpassing Claude 3 Opus on the leaderboard, and a 36.5 win rate on Arena-Hard -- making it the strongest 8B open-source model.