Recently, large language models (LLM) based on transformers are facing memory bottleneck issues due to KV cache, especially in long sequence handling. Previous researches proposed KV cache compression techniques that identify insignificant tokens based on Accumulative Attention Scores and removes their items from KV cache, noting that only few tokens play an important role in attention operations. However, we have observed that the existing Accumulative Attention Score is not suitable for the transformer decoder structure. In the decoder model, the number of times the Attention Score accumulates varies depending on the order of token appearance due to the effect of masking, causing an uneven comparison between tokens. To solve this, we propose Accumulative Attention Score with Forgetting Factor (A2SF) technique, which introduces a "Forgetting Factor" in the Attention Score accumulation process. A2SF applies a penalty to the past Attention Score generated from old tokens by repeatedly multiplying the Forgetting Factor to the Attention Score over time. Therefore, older tokens receive a larger penalty, providing fairness among different ages of tokens. Through the fair comparison among tokens, we can more effectively select important tokens. We have verified the accuracy improvement through A2SF in the OPT and LLaMA models and A2SF improves the accuracy of LLaMA 2 by up to 7.8% and 5.1% on 1-shot and 0-shot.

In contrast, LLMs, with their ability to understand nuances of language and context, offer a promising solution The rapid evolution of large language models (LLMs) has opened to overcome these limitations. However, a crucial prerequisite up new possibilities for applications such as context-driven product for LLMs to excel in product recommendation is their possession of recommendations. However, the effectiveness of these models in comprehensive knowledge about the entire inventory of products this context is heavily reliant on their comprehensive understanding available for sale. of the product inventory. This paper presents a novel approach In this paper, we propose a novel approach to equip LLMs with to equipping LLMs with product knowledge by training them to respond product knowledge by training them to generate contextual responses contextually to synthetic search queries that include product to synthetic search queries containing product IDs.

Adapting questionnaires to new languages is a resource-intensive process often requiring the hiring of multiple independent translators, which limits the ability of researchers to conduct cross-cultural research and effectively creates inequalities in research and society. This work presents a prototype tool that can expedite the questionnaire translation process. The tool incorporates forward-backward translation using DeepL alongside GPT-4-generated translation quality evaluations and improvement suggestions. We conducted two online studies in which participants translated questionnaires from English to either German (Study 1; n=10) or Portuguese (Study 2; n=20) using our prototype. To evaluate the quality of the translations created using the tool, evaluation scores between conventionally translated and tool-supported versions were compared. Our results indicate that integrating LLM-generated translation quality evaluations and suggestions for improvement can help users independently attain results similar to those provided by conventional, non-NLP-supported translation methods. This is the first step towards more equitable questionnaire-based research, powered by AI.

This research paper contributes to the computing education research community's understanding of Generative AI (GenAI) in the context of introductory programming, and specifically, how students utilize related tools, such as ChatGPT. An increased understanding of students' use is mandatory for educators and higher education institutions, as GenAI is here to stay, and its performance is likely to improve rapidly in the near future. Learning about students' use patterns is not only crucial to support their learning, but to develop adequate forms of instruction and assessment. With the rapid advancement of AI, its broad availability, and ubiquitous presence in educational environments, elaborating how AI can enhance learning experiences, especially in courses such as introductory programming is important. To date, most studies have focused on the educator's perspective on GenAI, its performance, characteristics, and limitations. However, the student perspective, and how they actually use GenAI tools in course contexts, has not been subject to a great number of studies. Therefore, this study is guided by the following research questions: (1) What do students report on their use pattern of ChatGPT in the context of introductory programming exercises? and (2) How do students perceive ChatGPT in the context of introductory programming exercises? To address these questions, computing students at a large German university were asked to solve programming tasks with the assistance of ChatGPT as part of their introductory programming course. Students (n=298) provided information regarding the use of ChatGPT, and their evaluation of the tool via an online survey. This research provides a comprehensive evaluation of ChatGPT-3.5's application by novice programmers in a higher education context...

We describe the development and capabilities of Meltemi 7B, the first open Large Language Model for the Greek language. Meltemi 7B has 7 billion parameters and is trained on a 40 billion token Greek corpus. For the development of Meltemi 7B, we adapt Mistral, by continuous pretraining on the Greek Corpus. Meltemi 7B contains up-to-date information up to September 2023. Furthermore, we have translated and curated a Greek instruction corpus, which has been used for the instruction-tuning of a chat model, named Meltemi 7B Instruct. Special care has been given to the alignment and the removal of toxic content for the Meltemi 7B Instruct. The developed models are evaluated on a broad set of collected evaluation corpora, and examples of prompts and responses are presented. Both Meltemi 7B and Meltemi 7B Instruct are available at https://huggingface.co/ilsp under the Apache 2.0 license.

The integration of artificial intelligence into agricultural practices, specifically through Consultation on Intelligent Agricultural Machinery Management (CIAMM), has the potential to revolutionize efficiency and sustainability in farming. This paper introduces a novel approach that leverages large language models (LLMs), particularly GPT-4, combined with multi-round prompt engineering to enhance decision-making processes in agricultural machinery management. We systematically developed and refined prompts to guide the LLMs in generating precise and contextually relevant outputs. Our approach was evaluated using a manually curated dataset from various online sources, and performance was assessed with accuracy and GPT-4 Scores. Comparative experiments were conducted using LLama-2-70B, ChatGPT, and GPT-4 models, alongside baseline and state-of-the-art methods such as Chain of Thought (CoT) and Thought of Thought (ThoT). The results demonstrate that our method significantly outperforms these approaches, achieving higher accuracy and relevance in generated responses. This paper highlights the potential of advanced prompt engineering techniques in improving the robustness and applicability of AI in agricultural contexts.

Visual question answering (VQA) is the task of providing accurate answers to natural language questions based on visual input. Programmatic VQA (PVQA) models have been gaining attention recently. These use large language models (LLMs) to formulate executable programs that address questions requiring complex visual reasoning. However, there are challenges in enabling LLMs to comprehend the usage of image processing modules and generate relevant code. To overcome these challenges, this paper introduces PyramidCoder, a novel prompting framework for PVQA models. PyramidCoder consists of three hierarchical levels, each serving a distinct purpose: query rephrasing, code generation, and answer aggregation. Notably, PyramidCoder utilizes a single frozen LLM and pre-defined prompts at each level, eliminating the need for additional training and ensuring flexibility across various LLM architectures. Compared to the state-of-the-art PVQA model, our approach improves accuracy by at least 0.5% on the GQA dataset, 1.4% on the VQAv2 dataset, and 2.9% on the NLVR2 dataset.

Modern language models (LMs) pose a new challenge in capability assessment. Static benchmarks inevitably saturate without providing confidence in the deployment tolerances of LM-based systems, but developers nonetheless claim that their models have generalized traits such as reasoning or open-domain language understanding based on these flawed metrics. The science and practice of LMs requires a new approach to benchmarking which measures specific capabilities with dynamic assessments. To be confident in our metrics, we need a new discipline of model metrology -- one which focuses on how to generate benchmarks that predict performance under deployment. Motivated by our evaluation criteria, we outline how building a community of model metrology practitioners -- one focused on building tools and studying how to measure system capabilities -- is the best way to meet these needs to and add clarity to the AI discussion.

Track finding in particle data is a challenging pattern recognition problem in High Energy Physics. It takes as inputs a point cloud of space points and labels them so that space points created by the same particle have the same label. The list of space points with the same label is a track candidate. We argue that this pattern recognition problem can be formulated as a sorting problem, of which the inputs are a list of space points sorted by their distances away from the collision points and the outputs are the space points sorted by their labels. In this paper, we propose the TrackSorter algorithm: a Transformer-based algorithm for pattern recognition in particle data. TrackSorter uses a simple tokenization scheme to convert space points into discrete tokens. It then uses the tokenized space points as inputs and sorts the input tokens into track candidates. TrackSorter is a novel end-to-end track finding algorithm that leverages Transformer-based models to solve pattern recognition problems. It is evaluated on the TrackML dataset and has good track finding performance.

How do language models learn to make predictions during pre-training? To study this, we extract learning curves from five autoregressive English language model pre-training runs, for 1M unseen tokens in context. We observe that the language models generate short repetitive phrases before learning to generate longer and more coherent text. We also find that individual tokens often exhibit sudden increases or decreases in loss that are surprisingly consistent across pre-training runs. To better understand these fluctuations, we quantify the final surprisal, within-run variability, age of acquisition, forgettability, and cross-run variability of learning curves for individual tokens in context. More frequent tokens reach lower final surprisals, exhibit less variability within and across pre-training runs, are learned earlier, and are less likely to be "forgotten" during pre-training. Higher n-gram probabilities further accentuate these effects. Independent of the target token, shorter and more frequent contexts correlate with marginally more stable and quickly acquired predictions. Based on our results, we argue for the existence of sequential learning dependencies between different model capabilities, and we characterize language model learning as early n-gram learning before gradual refinement of tail n-gram predictions.