This paper addresses a new motion planning problem for mobile robots tasked with accomplishing multiple high-level sub-tasks, expressed using natural language (NL). These sub-tasks should be accomplished in a temporal and logical order. To formally define the overarching mission, we leverage Linear Temporal Logic (LTL) defined over atomic predicates modeling these NL-based sub-tasks. This is in contrast to related planning approaches that define LTL tasks over atomic predicates capturing desired low-level system configurations. Our goal is to design robot plans that satisfy LTL tasks defined over NL-based atomic propositions. A novel technical challenge arising in this setup lies in reasoning about correctness of a robot plan with respect to such LTL-encoded tasks. To address this problem, we propose HERACLEs, a hierarchical conformal natural language planner, that relies on (i) automata theory to determine what NL-specified sub-tasks should be accomplished next to make mission progress; (ii) Large Language Models to design robot plans satisfying these sub-tasks; and (iii) conformal prediction to reason probabilistically about correctness of the designed plans and to determine if external assistance is required. We provide theoretical probabilistic mission satisfaction guarantees as well as extensive comparative experiments on mobile manipulation tasks.

Sentence Representation Learning (SRL) is a fundamental task in Natural Language Processing (NLP), with the Contrastive Learning of Sentence Embeddings (CSE) being the mainstream technique due to its superior performance. An intriguing phenomenon in CSE is the significant performance gap between supervised and unsupervised methods, with their only difference lying in the training data. Previous works attribute this performance gap to differences in two representation properties (alignment and uniformity). However, since alignment and uniformity only measure the results, they fail to answer "What aspects of the training data contribute to the performance gap?" and "How can the performance gap be narrowed?", In this paper, we conduct empirical experiments to answer these "What" and "How" questions. We first answer the "What" question by thoroughly comparing the behavior of supervised and unsupervised CSE during their respective training processes. From the comparison, we identify the similarity pattern as a key factor to the performance gap, and introduce a metric, called Relative Fitting Difficulty (RFD), to measure the complexity of the similarity pattern. Then, based on the insights gained from the "What" question, we tackle the "How" question by increasing the pattern complexity of the training data. We achieve this by leveraging the In-Context Learning (ICL) capability of the Large Language Model (LLM) to generate data that simulates complex patterns. By utilizing the hierarchical patterns in the LLM-generated data, we effectively narrow the gap between supervised and unsupervised CSE. We release our codes and appendix at https://github.com/BDBC-KG-NLP/NGCSE.

We present SeaEval, a benchmark for multilingual foundation models. In addition to characterizing how these models understand and reason with natural language, we also investigate how well they comprehend cultural practices, nuances, and values. Alongside standard accuracy metrics, we investigate the brittleness of foundation models in the dimensions of semantics and multilinguality. Our analyses span both open-sourced and closed models, leading to empirical results across classic NLP tasks, reasoning, and cultural comprehension. Key findings indicate (1) Most models exhibit varied behavior when given paraphrased instructions. (2) Many models still suffer from exposure bias (e.g., positional bias, majority label bias). (3) For questions rooted in factual, scientific, and commonsense knowledge, consistent responses are expected across multilingual queries that are semantically equivalent. Yet, most models surprisingly demonstrate inconsistent performance on these queries. (4) Multilingually-trained models have not attained "balanced multilingual" capabilities. Our endeavors underscore the need for more generalizable semantic representations and enhanced multilingual contextualization. SeaEval can serve as a launchpad for more thorough investigations and evaluations for multilingual and multicultural scenarios.

Software engineering is a domain characterized by intricate decision-making processes, often relying on nuanced intuition and consultation. Recent advancements in deep learning have started to revolutionize software engineering practices through elaborate designs implemented at various stages of software development. In this paper, we present an innovative paradigm that leverages large language models (LLMs) throughout the entire software development process, streamlining and unifying key processes through natural language communication, thereby eliminating the need for specialized models at each phase. At the core of this paradigm lies ChatDev, a virtual chat-powered software development company that mirrors the established waterfall model, meticulously dividing the development process into four distinct chronological stages: designing, coding, testing, and documenting. Each stage engages a team of "software agents", such as programmers, code reviewers, and test engineers, fostering collaborative dialogue and facilitating a seamless workflow. The chat chain acts as a facilitator, breaking down each stage into atomic subtasks. This enables dual roles, allowing for proposing and validating solutions through context-aware communication, leading to efficient resolution of specific subtasks. The instrumental analysis of ChatDev highlights its remarkable efficacy in software generation, enabling the completion of the entire software development process in under seven minutes at a cost of less than one dollar. It not only identifies and alleviates potential vulnerabilities but also rectifies potential hallucinations while maintaining commendable efficiency and cost-effectiveness. The potential of ChatDev unveils fresh possibilities for integrating LLMs into the realm of software development. Our code is available at https://github.com/OpenBMB/ChatDev.

Controllable text generation is a challenging and meaningful field in natural language generation (NLG). Especially, poetry generation is a typical one with well-defined and strict conditions for text generation which is an ideal playground for the assessment of current methodologies. While prior works succeeded in controlling either semantic or metrical aspects of poetry generation, simultaneously addressing both remains a challenge. In this paper, we pioneer the use of the Diffusion model for generating sonnets and Chinese SongCi poetry to tackle such challenges. In terms of semantics, our PoetryDiffusion model, built upon the Diffusion model, generates entire sentences or poetry by comprehensively considering the entirety of sentence information. This approach enhances semantic expression, distinguishing it from autoregressive and large language models (LLMs). For metrical control, the separation feature of diffusion generation and its constraint control module enable us to flexibly incorporate a novel metrical controller to manipulate and evaluate metrics (format and rhythm). The denoising process in PoetryDiffusion allows for gradual enhancement of semantics and flexible integration of the metrical controller which can calculate and impose penalties on states that stray significantly from the target control distribution. Experimental results on two datasets demonstrate that our model outperforms existing models in automatic evaluation of semantic, metrical, and overall performance as well as human evaluation.

In-context learning (ICL) emerges as a promising capability of large language models (LLMs) by providing them with demonstration examples to perform diverse tasks. However, the underlying mechanism of how LLMs learn from the provided context remains under-explored. In this paper, we investigate the working mechanism of ICL through an information flow lens. Our findings reveal that label words in the demonstration examples function as anchors: (1) semantic information aggregates into label word representations during the shallow computation layers' processing; (2) the consolidated information in label words serves as a reference for LLMs' final predictions. Based on these insights, we introduce an anchor re-weighting method to improve ICL performance, a demonstration compression technique to expedite inference, and an analysis framework for diagnosing ICL errors in GPT2-XL. The promising applications of our findings again validate the uncovered ICL working mechanism and pave the way for future studies.

Language tasks involving character-level manipulations (e.g., spelling corrections, arithmetic operations, word games) are challenging for models operating on subword units. To address this, we develop a causal intervention framework to learn robust and interpretable character representations inside subword-based language models. Our method treats each character as a typed variable in a causal model and learns such causal structures by adapting the interchange intervention training method of Geiger et al. (2021). We additionally introduce a suite of character-level tasks that systematically vary in their dependence on meaning and sequence-level context. While character-level models still perform best on purely form-based tasks like string reversal, our method outperforms character-level models on more complex tasks that blend form, meaning, and context, such as spelling correction in context and word search games. Compared with standard subword-based models, our approach also significantly improves robustness on unseen token sequences and leads to human-interpretable internal representations of characters.

In this note, I introduce Estimated Performance Rating (PR$^e$), a novel system for evaluating player performance in sports and games. PR$^e$ addresses a key limitation of the Tournament Performance Rating (TPR) system, which is undefined for zero or perfect scores in a series of games. PR$^e$ is defined as the rating that solves an optimization problem related to scoring probability, making it applicable for any performance level. The main theorem establishes that the PR$^e$ of a player is equivalent to the TPR whenever the latter is defined. I then apply this system to historically significant win-streaks in association football, tennis, and chess. Beyond sports, PR$^e$ has broad applicability in domains where Elo ratings are used, from college rankings to the evaluation of large language models.

The evolution of autonomous driving has made remarkable advancements in recent years, evolving into a tangible reality. However, a human-centric large-scale adoption hinges on meeting a variety of multifaceted requirements. To ensure that the autonomous system meets the user's intent, it is essential to accurately discern and interpret user commands, especially in complex or emergency situations. To this end, we propose to leverage the reasoning capabilities of Large Language Models (LLMs) to infer system requirements from in-cabin users' commands. Through a series of experiments that include different LLM models and prompt designs, we explore the few-shot multivariate binary classification accuracy of system requirements from natural language textual commands. We confirm the general ability of LLMs to understand and reason about prompts but underline that their effectiveness is conditioned on the quality of both the LLM model and the design of appropriate sequential prompts. Code and models are public with the link \url{https://github.com/KTH-RPL/DriveCmd_LLM}.

Modeling structure and behavior of software systems plays a crucial role, in various areas of software engineering. As with other software engineering artifacts, software models are subject to evolution. Supporting modelers in evolving models by model completion facilities and providing high-level edit operations such as frequently occurring editing patterns is still an open problem. Recently, large language models (i.e., generative neural networks) have garnered significant attention in various research areas, including software engineering. In this paper, we explore the potential of large language models in supporting the evolution of software models in software engineering. We propose an approach that utilizes large language models for model completion and discovering editing patterns in model histories of software systems. Through controlled experiments using simulated model repositories, we conduct an evaluation of the potential of large language models for these two tasks. We have found that large language models are indeed a promising technology for supporting software model evolution, and that it is worth investigating further in the area of software model evolution.