Modern code review is a process in which an incremental code contribution made by a code author is reviewed by one or more peers before it is committed to the version control system. An important element of modern code review is verifying that code contributions adhere to best practices. While some of these best practices can be automatically verified, verifying others is commonly left to human reviewers. This paper reports on the development, deployment, and evaluation of AutoCommenter, a system backed by a large language model that automatically learns and enforces coding best practices. We implemented AutoCommenter for four programming languages (C++, Java, Python, and Go) and evaluated its performance and adoption in a large industrial setting. Our evaluation shows that an end-to-end system for learning and enforcing coding best practices is feasible and has a positive impact on the developer workflow. Additionally, this paper reports on the challenges associated with deploying such a system to tens of thousands of developers and the corresponding lessons learned.
Autonomous driving is a complex task which requires advanced decision making and control algorithms. Understanding the rationale behind the autonomous vehicles' decision is crucial to ensure their safe and effective operation on highway driving. This study presents a novel approach, HighwayLLM, which harnesses the reasoning capabilities of large language models (LLMs) to predict the future waypoints for ego-vehicle's navigation. Our approach also utilizes a pre-trained Reinforcement Learning (RL) model to serve as a high-level planner, making decisions on appropriate meta-level actions. The HighwayLLM combines the output from the RL model and the current state information to make safe, collision-free, and explainable predictions for the next states, thereby constructing a trajectory for the ego-vehicle. Subsequently, a PID-based controller guides the vehicle to the waypoints predicted by the LLM agent. This integration of LLM with RL and PID enhances the decision-making process and provides interpretability for highway autonomous driving.
The reliance of language model training on massive amounts of computation and vast datasets scraped from potentially low-quality, copyrighted, or sensitive data has come into question practically, legally, and ethically. Federated learning provides a plausible alternative by enabling previously untapped data to be voluntarily gathered from collaborating organizations. However, when scaled globally, federated learning requires collaboration across heterogeneous legal, security, and privacy regimes while accounting for the inherent locality of language data; this further exacerbates the established challenge of federated statistical heterogeneity. We propose a Worldwide Federated Language Model Training~(WorldLM) system based on federations of federations, where each federation has the autonomy to account for factors such as its industry, operating jurisdiction, or competitive environment. WorldLM enables such autonomy in the presence of statistical heterogeneity via partial model localization by allowing sub-federations to attentively aggregate key layers from their constituents. Furthermore, it can adaptively share information across federations via residual layer embeddings. Evaluations of language modeling on naturally heterogeneous datasets show that WorldLM outperforms standard federations by up to $1.91\times$, approaches the personalized performance of fully local models, and maintains these advantages under privacy-enhancing techniques.
This paper describes a rapid feasibility study of using GPT-4, a large language model (LLM), to (semi)automate data extraction in systematic reviews. Despite the recent surge of interest in LLMs there is still a lack of understanding of how to design LLM-based automation tools and how to robustly evaluate their performance. During the 2023 Evidence Synthesis Hackathon we conducted two feasibility studies. Firstly, to automatically extract study characteristics from human clinical, animal, and social science domain studies. We used two studies from each category for prompt-development; and ten for evaluation. Secondly, we used the LLM to predict Participants, Interventions, Controls and Outcomes (PICOs) labelled within 100 abstracts in the EBM-NLP dataset. Overall, results indicated an accuracy of around 80%, with some variability between domains (82% for human clinical, 80% for animal, and 72% for studies of human social sciences). Causal inference methods and study design were the data extraction items with the most errors. In the PICO study, participants and intervention/control showed high accuracy (>80%), outcomes were more challenging. Evaluation was done manually; scoring methods such as BLEU and ROUGE showed limited value. We observed variability in the LLMs predictions and changes in response quality. This paper presents a template for future evaluations of LLMs in the context of data extraction for systematic review automation. Our results show that there might be value in using LLMs, for example as second or third reviewers. However, caution is advised when integrating models such as GPT-4 into tools. Further research on stability and reliability in practical settings is warranted for each type of data that is processed by the LLM.
The objective of this paper is to predict (A) whether a sentence in a written text expresses an emotion, (B) the mode(s) in which it is expressed, (C) whether it is basic or complex, and (D) its emotional category. One of our major contributions, through a dataset and a model, is to integrate the fact that an emotion can be expressed in different modes: from a direct mode, essentially lexicalized, to a more indirect mode, where emotions will only be suggested, a mode that NLP approaches generally don't take into account. Another originality is that the scope is on written texts, as opposed usual work focusing on conversational (often multi-modal) data. In this context, modes of expression are seen as a factor towards the automatic analysis of complexity in texts. Experiments on French texts show acceptable results compared to the human annotators' agreement, and outperforming results compared to using a large language model with in-context learning (i.e. no fine-tuning).
The absence of adequately sufficient expert-level tumor annotations hinders the effectiveness of supervised learning based opportunistic cancer screening on medical imaging. Clinical reports (that are rich in descriptive textual details) can offer a "free lunch'' supervision information and provide tumor location as a type of weak label to cope with screening tasks, thus saving human labeling workloads, if properly leveraged. However, predicting cancer only using such weak labels can be very changeling since tumors are usually presented in small anatomical regions compared to the whole 3D medical scans. Weakly semi-supervised learning (WSSL) utilizes a limited set of voxel-level tumor annotations and incorporates alongside a substantial number of medical images that have only off-the-shelf clinical reports, which may strike a good balance between minimizing expert annotation workload and optimizing screening efficacy. In this paper, we propose a novel text-guided learning method to achieve highly accurate cancer detection results. Through integrating diagnostic and tumor location text prompts into the text encoder of a vision-language model (VLM), optimization of weakly supervised learning can be effectively performed in the latent space of VLM, thereby enhancing the stability of training. Our approach can leverage clinical knowledge by large-scale pre-trained VLM to enhance generalization ability, and produce reliable pseudo tumor masks to improve cancer detection. Our extensive quantitative experimental results on a large-scale cancer dataset, including 1,651 unique patients, validate that our approach can reduce human annotation efforts by at least 70% while maintaining comparable cancer detection accuracy to competing fully supervised methods (AUC value 0.961 versus 0.966).
Multimodal Large Language Models (MLLMs) are widely regarded as crucial in the exploration of Artificial General Intelligence (AGI). The core of MLLMs lies in their capability to achieve cross-modal alignment. To attain this goal, current MLLMs typically follow a two-phase training paradigm: the pre-training phase and the instruction-tuning phase. Despite their success, there are shortcomings in the modeling of alignment capabilities within these models. Firstly, during the pre-training phase, the model usually assumes that all image-text pairs are uniformly aligned, but in fact the degree of alignment between different image-text pairs is inconsistent. Secondly, the instructions currently used for finetuning incorporate a variety of tasks and different tasks usually require different levels of alignment capabilities, but previous MLLMs overlook these differentiated alignment needs. To tackle these issues, we propose a new multimodal large language model AlignGPT. In the pre-training stage, instead of treating all image-text pairs equally, we divide them into different groups according to the degrees of alignment of them. Then, the model is trained to learn the representations of different alignment levels. In the instruction-tuning phase, we adaptively combine these representations of alignment levels to meet the dynamic alignment needs of different tasks. Extensive experimental results show that our model achieves competitive performance on 12 benchmarks.
This paper introduces distributed speculative inference (DSI), a novel inference algorithm that is provably faster than speculative inference (SI) [leviathan2023, chen2023, miao2024, sun2025, timor2025] and standard autoregressive inference (non-SI). Like other SI algorithms, DSI operates on frozen language models (LMs), requiring no training or architectural modifications, and it preserves the target distribution. Prior studies on SI have demonstrated empirical speedups over non-SI--but rely on sufficiently fast and accurate drafters, which are often unavailable in practice. We identify a gap where SI can be slower than non-SI if drafters are too slow or inaccurate. We close this gap by proving that DSI is faster than both SI and non-SI--given any drafters. DSI is therefore not only faster than SI, but also unlocks the acceleration of LMs for which SI fails. DSI leverages speculation parallelism (SP), a novel type of task parallelism, to orchestrate target and drafter instances that overlap in time, establishing a new foundational tradeoff between computational resources and latency. Our simulations show that DSI is 1.29-1.92x faster than SI in single-node setups for various off-the-shelf LMs and tasks. We open-source all our code.
We present ChatScene, a Large Language Model (LLM)-based agent that leverages the capabilities of LLMs to generate safety-critical scenarios for autonomous vehicles. Given unstructured language instructions, the agent first generates textually described traffic scenarios using LLMs. These scenario descriptions are subsequently broken down into several sub-descriptions for specified details such as behaviors and locations of vehicles. The agent then distinctively transforms the textually described sub-scenarios into domain-specific languages, which then generate actual code for prediction and control in simulators, facilitating the creation of diverse and complex scenarios within the CARLA simulation environment. A key part of our agent is a comprehensive knowledge retrieval component, which efficiently translates specific textual descriptions into corresponding domain-specific code snippets by training a knowledge database containing the scenario description and code pairs. Extensive experimental results underscore the efficacy of ChatScene in improving the safety of autonomous vehicles. For instance, the scenarios generated by ChatScene show a 15% increase in collision rates compared to state-of-the-art baselines when tested against different reinforcement learning-based ego vehicles. Furthermore, we show that by using our generated safety-critical scenarios to fine-tune different RL-based autonomous driving models, they can achieve a 9% reduction in collision rates, surpassing current SOTA methods. ChatScene effectively bridges the gap between textual descriptions of traffic scenarios and practical CARLA simulations, providing a unified way to conveniently generate safety-critical scenarios for safety testing and improvement for AVs.
Generating personalized 3D avatars is crucial for AR/VR. However, recent text-to-3D methods that generate avatars for celebrities or fictional characters, struggle with everyday people. Methods for faithful reconstruction typically require full-body images in controlled settings. What if a user could just upload their personal "OOTD" (Outfit Of The Day) photo collection and get a faithful avatar in return? The challenge is that such casual photo collections contain diverse poses, challenging viewpoints, cropped views, and occlusion (albeit with a consistent outfit, accessories and hairstyle). We address this novel "Album2Human" task by developing PuzzleAvatar, a novel model that generates a faithful 3D avatar (in a canonical pose) from a personal OOTD album, while bypassing the challenging estimation of body and camera pose. To this end, we fine-tune a foundational vision-language model (VLM) on such photos, encoding the appearance, identity, garments, hairstyles, and accessories of a person into (separate) learned tokens and instilling these cues into the VLM. In effect, we exploit the learned tokens as "puzzle pieces" from which we assemble a faithful, personalized 3D avatar. Importantly, we can customize avatars by simply inter-changing tokens. As a benchmark for this new task, we collect a new dataset, called PuzzleIOI, with 41 subjects in a total of nearly 1K OOTD configurations, in challenging partial photos with paired ground-truth 3D bodies. Evaluation shows that PuzzleAvatar not only has high reconstruction accuracy, outperforming TeCH and MVDreamBooth, but also a unique scalability to album photos, and strong robustness. Our code and data are publicly available for research purpose at https://puzzleavatar.is.tue.mpg.de/
The spatiotemporal data generated by massive sensors in the Internet of Things (IoT) is extremely dynamic, heterogeneous, large scale and time-dependent. It poses great challenges (e.g. accuracy, reliability, and stability) in real-time analysis and decision making for different IoT applications. The complexity of IoT data prevents the common people from gaining a deeper understanding of it. Agentized systems help address the lack of data insight for the common people. We propose a generic framework, namely CityGPT, to facilitate the learning and analysis of IoT time series with an end-to-end paradigm. CityGPT employs three agents to accomplish the spatiotemporal analysis of IoT data. The requirement agent facilitates user inputs based on natural language. Then, the analysis tasks are decomposed into temporal and spatial analysis processes, completed by corresponding data analysis agents (temporal and spatial agents). Finally, the spatiotemporal fusion agent visualizes the system's analysis results by receiving analysis results from data analysis agents and invoking sub-visualization agents, and can provide corresponding textual descriptions based on user demands. To increase the insight for common people using our framework, we have agnentized the framework, facilitated by a large language model (LLM), to increase the data comprehensibility. Our evaluation results on real-world data with different time dependencies show that the CityGPT framework can guarantee robust performance in IoT computing.
Generative modeling over discrete data has recently seen numerous success stories, with applications spanning language modeling, biological sequence design, and graph-structured molecular data. The predominant generative modeling paradigm for discrete data is still autoregressive, with more recent alternatives based on diffusion or flow-matching falling short of their impressive performance in continuous data settings, such as image or video generation. In this work, we introduce Fisher-Flow, a novel flow-matching model for discrete data. Fisher-Flow takes a manifestly geometric perspective by considering categorical distributions over discrete data as points residing on a statistical manifold equipped with its natural Riemannian metric: the $\textit{Fisher-Rao metric}$. As a result, we demonstrate discrete data itself can be continuously reparameterised to points on the positive orthant of the $d$-hypersphere $\mathbb{S}^d_+$, which allows us to define flows that map any source distribution to target in a principled manner by transporting mass along (closed-form) geodesics of $\mathbb{S}^d_+$. Furthermore, the learned flows in Fisher-Flow can be further bootstrapped by leveraging Riemannian optimal transport leading to improved training dynamics. We prove that the gradient flow induced by Fisher-Flow is optimal in reducing the forward KL divergence. We evaluate Fisher-Flow on an array of synthetic and diverse real-world benchmarks, including designing DNA Promoter, and DNA Enhancer sequences. Empirically, we find that Fisher-Flow improves over prior diffusion and flow-matching models on these benchmarks.
In the expanding field of language model applications, medical knowledge representation remains a significant challenge due to the specialized nature of the domain. Large language models, such as GPT-4, obtain reasonable scores on medical question answering tasks, but smaller models are far behind. In this work, we introduce a method to improve the proficiency of a small language model in the medical domain by employing a two-fold approach. We first fine-tune the model on a corpus of medical textbooks. Then, we use GPT-4 to generate questions similar to the downstream task, prompted with textbook knowledge, and use them to fine-tune the model. Additionally, we introduce ECN-QA, a novel medical question answering dataset containing ``progressive questions'' composed of related sequential questions. We show the benefits of our training strategy on this dataset. The study's findings highlight the potential of small language models in the medical domain when appropriately fine-tuned. The code and weights are available at https://github.com/raidium-med/MQG.
Preference optimization is a standard approach to fine-tuning large language models to align with human preferences. The quality, diversity, and quantity of the preference dataset are critical to the effectiveness of preference optimization. However, obtaining a large amount of high-quality and diverse preference annotations is difficult in many applications. This raises the question of how to use the limited annotation budget to create an effective preference dataset. To this end, we propose Annotation-Efficient Preference Optimization (AEPO). Instead of exhaustively annotating preference over all available response texts, AEPO selects a subset of responses that maximizes quality and diversity from the available responses, and then annotates preference over the selected ones. In this way, AEPO focuses the annotation budget on labeling preference over a smaller subset of responses with diversity and of high quality. We evaluate the performance of Direct Preference Optimization (DPO) using AEPO and show that it outperforms models trained using a standard DPO with the same annotation budget. Our code is available at https://github.com/CyberAgentAILab/annotation-efficient-po
While there have been extensive studies in code generation by large language models (LLM), where benchmarks like HumanEval have been surpassed with an impressive 96.3% success rate, these benchmarks predominantly judge a model's performance on basic function-level code generation and lack the critical thinking and concept of scope required of real-world scenarios such as solving GitHub issues. This research introduces the application of the Tree of Thoughts (ToT) language model reasoning framework for enhancing the decision-making and problem-solving abilities of LLMs for this complex task. Compared to traditional input-output (IO) prompting and Retrieval Augmented Generation (RAG) techniques, ToT is designed to improve performance by facilitating a structured exploration of multiple reasoning trajectories and enabling self-assessment of potential solutions. We experimentally deploy ToT in tackling a Github issue contained within an instance of the SWE-bench. However, our results reveal that the ToT framework alone is not enough to give LLMs the critical reasoning capabilities to outperform existing methods. In this paper we analyze the potential causes of these shortcomings and identify key areas for improvement such as deepening the thought process and introducing agentic capabilities. The insights of this research are aimed at informing future directions for refining the application of ToT and better harnessing the potential of LLMs in real-world problem-solving scenarios.
We present the first domain-adapted and fully-trained large language model, RecGPT-7B, and its instruction-following variant, RecGPT-7B-Instruct, for text-based recommendation. Experimental results on rating prediction and sequential recommendation tasks show that our model, RecGPT-7B-Instruct, outperforms previous strong baselines. We are releasing our RecGPT models as well as their pre-training and fine-tuning datasets to facilitate future research and downstream applications in text-based recommendation. Public "huggingface" links to our RecGPT models and datasets are available at: https://github.com/VinAIResearch/RecGPT
Large language models LLMs have transformed AI and achieved breakthrough performance on a wide range of tasks In science the most interesting application of LLMs is for hypothesis formation A feature of LLMs which results from their probabilistic structure is that the output text is not necessarily a valid inference from the training text These are termed hallucinations and are harmful in many applications In science some hallucinations may be useful novel hypotheses whose validity may be tested by laboratory experiments Here we experimentally test the application of LLMs as a source of scientific hypotheses using the domain of breast cancer treatment We applied the LLM GPT4 to hypothesize novel synergistic pairs of FDA-approved noncancer drugs that target the MCF7 breast cancer cell line relative to the nontumorigenic breast cell line MCF10A In the first round of laboratory experiments GPT4 succeeded in discovering three drug combinations out of twelve tested with synergy scores above the positive controls GPT4 then generated new combinations based on its initial results this generated three more combinations with positive synergy scores out of four tested We conclude that LLMs are a valuable source of scientific hypotheses.
Recent advances in generative language modeling applied to discrete speech tokens presented a new avenue for text-to-speech (TTS) synthesis. These speech language models (SLMs), similarly to their textual counterparts, are scalable, probabilistic, and context-aware. While they can produce diverse and natural outputs, they sometimes face issues such as unintelligibility and the inclusion of non-speech noises or hallucination. As the adoption of this innovative paradigm in speech synthesis increases, there is a clear need for an in-depth evaluation of its capabilities and limitations. In this paper, we evaluate TTS from a discrete token-based SLM, through both automatic metrics and listening tests. We examine five key dimensions: speaking style, intelligibility, speaker consistency, prosodic variation, spontaneous behaviour. Our results highlight the model's strength in generating varied prosody and spontaneous outputs. It is also rated higher in naturalness and context appropriateness in listening tests compared to a conventional TTS. However, the model's performance in intelligibility and speaker consistency lags behind traditional TTS. Additionally, we show that increasing the scale of SLMs offers a modest boost in robustness. Our findings aim to serve as a benchmark for future advancements in generative SLMs for speech synthesis.
Affective Image Manipulation (AIM) seeks to modify user-provided images to evoke specific emotional responses. This task is inherently complex due to its twofold objective: significantly evoking the intended emotion, while preserving the original image composition. Existing AIM methods primarily adjust color and style, often failing to elicit precise and profound emotional shifts. Drawing on psychological insights, we introduce EmoEdit, which extends AIM by incorporating content modifications to enhance emotional impact. Specifically, we first construct EmoEditSet, a large-scale AIM dataset comprising 40,120 paired data through emotion attribution and data construction. To make existing generative models emotion-aware, we design the Emotion adapter and train it using EmoEditSet. We further propose an instruction loss to capture the semantic variations in data pairs. Our method is evaluated both qualitatively and quantitatively, demonstrating superior performance compared to existing state-of-the-art techniques. Additionally, we showcase the portability of our Emotion adapter to other diffusion-based models, enhancing their emotion knowledge with diverse semantics.
Current methods for Video Moment Retrieval (VMR) struggle to align complex situations involving specific environmental details, character descriptions, and action narratives. To tackle this issue, we propose a Large Language Model-guided Moment Retrieval (LMR) approach that employs the extensive knowledge of Large Language Models (LLMs) to improve video context representation as well as cross-modal alignment, facilitating accurate localization of target moments. Specifically, LMR introduces a context enhancement technique with LLMs to generate crucial target-related context semantics. These semantics are integrated with visual features for producing discriminative video representations. Finally, a language-conditioned transformer is designed to decode free-form language queries, on the fly, using aligned video representations for moment retrieval. Extensive experiments demonstrate that LMR achieves state-of-the-art results, outperforming the nearest competitor by up to 3.28\% and 4.06\% on the challenging QVHighlights and Charades-STA benchmarks, respectively. More importantly, the performance gains are significantly higher for localization of complex queries.