Large Language Models (LLMs) have demonstrated proficiency in a wide array of natural language processing tasks. However, its effectiveness over discourse-level event relation extraction (ERE) tasks remains unexplored. In this paper, we assess the effectiveness of LLMs in addressing discourse-level ERE tasks characterized by lengthy documents and intricate relations encompassing coreference, temporal, causal, and subevent types. Evaluation is conducted using an commercial model, GPT-3.5, and an open-source model, LLaMA-2. Our study reveals a notable underperformance of LLMs compared to the baseline established through supervised learning. Although Supervised Fine-Tuning (SFT) can improve LLMs performance, it does not scale well compared to the smaller supervised baseline model. Our quantitative and qualitative analysis shows that LLMs have several weaknesses when applied for extracting event relations, including a tendency to fabricate event mentions, and failures to capture transitivity rules among relations, detect long distance relations, or comprehend contexts with dense event mentions.

In this technical report, we present our solution for the EgoPlan Challenge in ICML 2024. To address the real-world egocentric task planning problem, we introduce a novel planning framework which comprises three stages: long-term memory Extraction, context-awared Planning, and multi-iteration Decision, named EPD. Given the task goal, task progress, and current observation, the extraction model first extracts task-relevant memory information from the progress video, transforming the complex long video into summarized memory information. The planning model then combines the context of the memory information with fine-grained visual information from the current observation to predict the next action. Finally, through multi-iteration decision-making, the decision model comprehensively understands the task situation and current state to make the most realistic planning decision. On the EgoPlan-Test set, EPD achieves a planning accuracy of 53.85% over 1,584 egocentric task planning questions. We have made all codes available at https://github.com/Kkskkkskr/EPD .

As large language models (LLMs) continue to evolve, the need for robust and standardized evaluation benchmarks becomes paramount. Evaluating the performance of these models is a complex challenge that requires careful consideration of various linguistic tasks, model architectures, and benchmarking methodologies. In recent years, various frameworks have emerged as noteworthy contributions to the field, offering comprehensive evaluation tests and benchmarks for assessing the capabilities of LLMs across diverse domains. This paper provides an exploration and critical analysis of some of these evaluation methodologies, shedding light on their strengths, limitations, and impact on advancing the state-of-the-art in natural language processing.

Visual question answering aims to provide responses to natural language questions given visual input. Recently, visual programmatic models (VPMs), which generate executable programs to answer questions through large language models (LLMs), have attracted research interest. However, they often require long input prompts to provide the LLM with sufficient API usage details to generate relevant code. To address this limitation, we propose AdaCoder, an adaptive prompt compression framework for VPMs. AdaCoder operates in two phases: a compression phase and an inference phase. In the compression phase, given a preprompt that describes all API definitions in the Python language with example snippets of code, a set of compressed preprompts is generated, each depending on a specific question type. In the inference phase, given an input question, AdaCoder predicts the question type and chooses the appropriate corresponding compressed preprompt to generate code to answer the question. Notably, AdaCoder employs a single frozen LLM and pre-defined prompts, negating the necessity of additional training and maintaining adaptability across different powerful black-box LLMs such as GPT and Claude. In experiments, we apply AdaCoder to ViperGPT and demonstrate that it reduces token length by 71.1%, while maintaining or even improving the performance of visual question answering.

Large language models (LLMs) have garnered increasing attention owing to their powerful logical reasoning capabilities. Generally, larger LLMs (L-LLMs) that require paid interfaces exhibit significantly superior performance compared to smaller LLMs (S-LLMs) that can be deployed on a variety of devices. Knowledge distillation (KD) aims to empower S-LLMs with the capabilities of L-LLMs, while S-LLMs merely mimic the outputs of L-LLMs, failing to get the powerful logical reasoning capabilities. Consequently, S-LLMs are helpless when it comes to planning and decision-making tasks that require logical reasoning capabilities. To tackle the identified challenges, we propose a novel framework called Logic Distillation (LD). Initially, LD employs L-LLMs to instantiate complex instructions into discrete functions and illustrates their usage to establish a function base. Subsequently, based on the function base, LD fine-tunes S-LLMs to learn the logic employed by L-LLMs in planning and decision-making. During testing, LD utilizes a retriever to identify the top-$K$ relevant functions based on instructions and current states, which will be selected and invoked by S-LLMs. Ultimately, S-LLMs yield planning and decision-making outcomes, function by function. Relevant experiments demonstrate that with the assistance of LD, S-LLMs can achieve outstanding results in planning and decision-making tasks, comparable to, or even surpassing, those of L-LLMs.

I bought an iPhone 15 the other day to replace my five-year-old iPhone 11. The phone is powered by the new A17 Pro chip and has a terabyte of data storage and accordingly was eye-wateringly expensive. I had, of course, finely honed rationales for splashing out on such a scale. I've always had a policy of writing only about kit that I buy with my own money (no freebies from tech companies), for example. The fancy A17 processor is needed to run the new "AI" stuff that Apple is promising to launch soon; the phone has a significantly better camera than my old handset had – which matters (to me) because my Substack blog goes out three times a week and I provide a new photograph for each edition; and, finally, a friend whose ancient iPhone is on its last legs might appreciate an iPhone 11 in good nick.

Freelancer has accused Anthropic, the AI startup behind the Claude large language models, of ignoring its "do not crawl" robots.txt Meanwhile, iFixit CEO Kyle Wiens said Anthropic has ignored the website's policy prohibiting the use of its content for AI model training. Matt Barrie, the chief executive of Freelancer, told The Information that Anthropic's ClaudeBot is "the most aggressive scraper by far." His website allegedly got 3.5 million visits from the company's crawler within a span of four hours, which is "probably about five times the volume of the number two" AI crawler. Similarly, Wiens posted on X/Twitter that Anthropic's bot hit iFixit's servers a million times in 24 hours.

China has once again extended its policy of censorship and surveillance as it looks to keep artificial intelligence (AI) models in check even as it races to advance the ever-expanding technology. The Chinese Communist Party (CCP) has introduced more regulative measures to make sure its home-based tech companies adhere to the party's ideological rules. All AI firms are required to participate in a government review which analyzes the companies' large language models (LLMs) to ensure they "embody core socialist values," as first reported by the Financial Times last week. A man walks past a photo of Chinese President Xi Jinping at the Museum of the Communist Party of China in Beijing on March 3, 2023. A NEW BREED OF MILITARY AI ROBO-DOGS COULD BE MARINES' NEW SECRET WEAPON China has long worked to suppress information accessible over the internet through the use of its "Great Firewall" -- which has been used to block a litany of items perceived as bad for the CCP, such as information surrounding the 1989 Tiananmen Square massacre or memes comparing Chinese President Xi Jinping to Winnie the Pooh.

The proliferation of large language models (LLMs) requires robust evaluation of their alignment with local values and ethical standards, especially as existing benchmarks often reflect the cultural, legal, and ideological values of their creators. \textsc{LocalValueBench}, introduced in this paper, is an extensible benchmark designed to assess LLMs' adherence to Australian values, and provides a framework for regulators worldwide to develop their own LLM benchmarks for local value alignment. Employing a novel typology for ethical reasoning and an interrogation approach, we curated comprehensive questions and utilized prompt engineering strategies to probe LLMs' value alignment. Our evaluation criteria quantified deviations from local values, ensuring a rigorous assessment process. Comparative analysis of three commercial LLMs by USA vendors revealed significant insights into their effectiveness and limitations, demonstrating the critical importance of value alignment. This study offers valuable tools and methodologies for regulators to create tailored benchmarks, highlighting avenues for future research to enhance ethical AI development.

Addressing school bullying effectively and promptly is crucial for the mental health of students. This study examined the potential of large language models (LLMs) to empower students by discerning between bullying and joking in school peer interactions. We employed ChatGPT-4, Gemini 1.5 Pro, and Claude 3 Opus, evaluating their effectiveness through human review. Our results revealed that not all LLMs were suitable for an agentic approach, with ChatGPT-4 showing the most promise. We observed variations in LLM outputs, possibly influenced by political overcorrectness, context window limitations, and pre-existing bias in their training data. ChatGPT-4 excelled in context-specific accuracy after implementing the agentic approach, highlighting its potential to provide continuous, real-time support to vulnerable students.