The advent of large language models (LLMs) has greatly facilitated code generation, but ensuring the functional correctness of generated code remains a challenge. Traditional validation methods are often time-consuming, error-prone, and impractical for large volumes of code. We introduce CodeSift, a novel framework that leverages LLMs as the first-line filter of code validation without the need for execution, reference code, or human feedback, thereby reducing the validation effort. We assess the effectiveness of our method across three diverse datasets encompassing two programming languages. Our results indicate that CodeSift outperforms state-of-the-art code evaluation methods. Internal testing conducted with subject matter experts reveals that the output generated by CodeSift is in line with human preference, reinforcing its effectiveness as a dependable automated code validation tool.

Rapid advancements have been made in extending Large Language Models (LLMs) to Large Multi-modal Models (LMMs). However, extending input modality of LLMs to video data remains a challenging endeavor, especially for long videos. Due to insufficient access to large-scale high-quality video data and the excessive compression of visual features, current methods exhibit limitations in effectively processing long videos. In this paper, we introduce Kangaroo, a powerful Video LMM aimed at addressing these challenges. Confronted with issue of inadequate training data, we develop a data curation system to build a large-scale dataset with high-quality annotations for vision-language pre-training and instruction tuning. In addition, we design a curriculum training pipeline with gradually increasing resolution and number of input frames to accommodate long videos. Evaluation results demonstrate that, with 8B parameters, Kangaroo achieves state-of-the-art performance across a variety of video understanding benchmarks while exhibiting competitive results on others. Particularly, on benchmarks specialized for long videos, Kangaroo excels some larger models with over 10B parameters and proprietary models.

The rapid obsolescence of information in Large Language Models (LLMs) has driven the development of various techniques to incorporate new facts. However, existing methods for knowledge editing still face difficulties with multi-hop questions that require accurate fact identification and sequential logical reasoning, particularly among numerous fact updates. To tackle these challenges, this paper introduces Graph Memory-based Editing for Large Language Models (GMeLLo), a straitforward and effective method that merges the explicit knowledge representation of Knowledge Graphs (KGs) with the linguistic flexibility of LLMs. Beyond merely leveraging LLMs for question answering, GMeLLo employs these models to convert free-form language into structured queries and fact triples, facilitating seamless interaction with KGs for rapid updates and precise multi-hop reasoning. Our results show that GMeLLo significantly surpasses current state-of-the-art knowledge editing methods in the multi-hop question answering benchmark, MQuAKE, especially in scenarios with extensive knowledge edits.

ABSTRACT This report analyzes the potential for large language models (LLMs) to expedite accurate replication of published message effects studies. We tested LLM-powered participants (personas) by replicating 133 experimental findings from 14 papers containing 45 recent studies in the Journal of Marketing (January 2023-May 2024). We used a new software tool, Viewpoints AI (https://viewpoints.ai/), that takes study designs, stimuli, and measures as input, automatically generates prompts for LLMs to act as a specified sample of unique personas, and collects their responses to produce a final output in the form of a complete dataset and statistical analysis. The underlying LLM used was Anthropic's Claude Sonnet 3.5. We generated 19,447 AI personas to replicate these studies with the exact same sample attributes, study designs, stimuli, and measures reported in the original human research. Our LLM replications successfully reproduced 76% of the original main effects (84 out of 111), demonstrating strong potential for AI-assisted replication of studies in which people respond to media stimuli. When including interaction effects, the overall replication rate was 68% (90 out of 133). The use of LLMs to replicate and accelerate marketing research on media effects is discussed with respect to the replication crisis in social science, potential solutions to generalizability problems in sampling subjects and experimental conditions, and the ability to rapidly test consumer responses to various media stimuli. We also address the limitations of this approach, particularly in replicating complex interaction effects in media response studies, and suggest areas for future research and improvement in AI-assisted experimental replication of media effects. STUDY OVERVIEW AND RELATED WORK Research about the effectiveness of media messages is increasingly difficult, attributable to both administrative challenges (e.g., stimulus acquisition and creation, data management demands of digital trace data, acquisition of participants and especially those in special groups like children, minorities and international groups), as well as requirements to deal with new and critical challenges to the very nature of social research, as exemplified by existential issues of replication and reproducibility, and the ability to generalize findings across people, media stimuli and experimental contexts. We briefly review these issues with an eye toward our current test of whether new LLM tools may help solve the problems mentioned, and with significant advantages in cost, time, and research personnel.

Reasoning about real-life events is a unifying challenge in AI and NLP that has profound utility in a variety of domains, while fallacy in high-stake applications could be catastrophic. Able to work with diverse text in these domains, large language models (LLMs) have proven capable of answering questions and solving problems. However, I show that end-to-end LLMs still systematically fail to reason about complex events, and they lack interpretability due to their black-box nature. To address these issues, I propose three general approaches to use LLMs in conjunction with a structured representation of events. The first is a language-based representation involving relations of sub-events that can be learned by LLMs via fine-tuning. The second is a semi-symbolic representation involving states of entities that can be predicted and leveraged by LLMs via few-shot prompting. The third is a fully symbolic representation that can be predicted by LLMs trained with structured data and be executed by symbolic solvers. On a suite of event reasoning tasks spanning common-sense inference and planning, I show that each approach greatly outperforms end-to-end LLMs with more interpretability. These results suggest manners of synergy between LLMs and structured representations for event reasoning and beyond.

With the strong representational power of large language models (LLMs), generative error correction (GER) for automatic speech recognition (ASR) aims to provide semantic and phonetic refinements to address ASR errors. This work explores how LLM-based GER can enhance and expand the capabilities of Japanese language processing, presenting the first GER benchmark for Japanese ASR with 0.9-2.6k text utterances. We also introduce a new multi-pass augmented generative error correction (MPA GER) by integrating multiple system hypotheses on the input side with corrections from multiple LLMs on the output side and then merging them. To the best of our knowledge, this is the first investigation of the use of LLMs for Japanese GER, which involves second-pass language modeling on the output transcriptions generated by the ASR system (e.g., N-best hypotheses). Our experiments demonstrated performance improvement in the proposed methods of ASR quality and generalization both in SPREDS-U1-ja and CSJ data.

The increasing sophistication of large language models (LLMs) has allowed for their more prominent presence in political science research. One particular area gathering significant attention in the field is the use of LLMs as annotators. Research has shown promising results, with LLMs often outperforming human coders (Gilardi, Alizadeh and Kubli, 2023) and providing comparable accuracy when labelling political text, across multiple languages (Heseltine and Clemm von Hohenberg, 2024). While researchers have evaluated the performance of LLMs as annotators across different domains, there still little information on how the known biases of LLMs (see Gallegos et al., 2024) can affect their performance. For human annotators, studies show that political cues, such as party, have an effect on their coding decisions (Laver and Garry, 2000; Benoit et al., 2016; Ennser-Jedenastik and Meyer, 2018).

We introduce the Logic-Enhanced Language Model Agents (LELMA) framework, a novel approach to enhance the trustworthiness of social simulations that utilize large language models (LLMs). While LLMs have gained attention as agents for simulating human behaviour, their applicability in this role is limited by issues such as inherent hallucinations and logical inconsistencies. LELMA addresses these challenges by integrating LLMs with symbolic AI, enabling logical verification of the reasoning generated by LLMs. This verification process provides corrective feedback, refining the reasoning output. The framework consists of three main components: an LLM-Reasoner for producing strategic reasoning, an LLM-Translator for mapping natural language reasoning to logic queries, and a Solver for evaluating these queries. This study focuses on decision-making in game-theoretic scenarios as a model of human interaction. Experiments involving the Hawk-Dove game, Prisoner's Dilemma, and Stag Hunt highlight the limitations of state-of-the-art LLMs, GPT-4 Omni and Gemini 1.0 Pro, in producing correct reasoning in these contexts. LELMA demonstrates high accuracy in error detection and improves the reasoning correctness of LLMs via self-refinement, particularly in GPT-4 Omni.

Generating high-quality shooting scripts containing information such as scene and shot language is essential for short drama script generation. We collect 6,660 popular short drama episodes from the Internet, each with an average of 100 short episodes, and the total number of short episodes is about 80,000, with a total duration of about 2,000 hours and totaling 10 terabytes (TB). We perform keyframe extraction and annotation on each episode to obtain about 10,000,000 shooting scripts. We perform 100 script restorations on the extracted shooting scripts based on our self-developed large short drama generation model SkyReels. This leads to a dataset containing 1,000,000,000 pairs of scripts and shooting scripts for short dramas, called SkyScript-100M. We compare SkyScript-100M with the existing dataset in detail and demonstrate some deeper insights that can be achieved based on SkyScript-100M. Based on SkyScript-100M, researchers can achieve several deeper and more far-reaching script optimization goals, which may drive a paradigm shift in the entire field of text-to-video and significantly advance the field of short drama video generation. The data and code are available at https://github.com/vaew/SkyScript-100M.

Speculative sampling has proven to be an effective solution to accelerate decoding from large language models, where the acceptance rate significantly determines the performance. Most previous works on improving the acceptance rate focus on aligned training and efficient decoding, implicitly paying less attention to the linkage of training and decoding. In this work, we first investigate the linkage of training and decoding for speculative sampling and then propose a solution named HArmonized Speculative Sampling (HASS). HASS improves the acceptance rate without extra inference overhead by harmonizing training and decoding on their objectives and contexts. Experiments on three LLaMA models demonstrate that HASS achieves 2.81x-3.65x