In this work, we revisit the problem of semi-supervised named entity recognition (NER) focusing on extremely light supervision, consisting of a lexicon containing only 10 examples per class. We introduce ELLEN, a simple, fully modular, neuro-symbolic method that blends fine-tuned language models with linguistic rules. These rules include insights such as ''One Sense Per Discourse'', using a Masked Language Model as an unsupervised NER, leveraging part-of-speech tags to identify and eliminate unlabeled entities as false negatives, and other intuitions about classifier confidence scores in local and global context. ELLEN achieves very strong performance on the CoNLL-2003 dataset when using the minimal supervision from the lexicon above. It also outperforms most existing (and considerably more complex) semi-supervised NER methods under the same supervision settings commonly used in the literature (i.e., 5% of the training data). Further, we evaluate our CoNLL-2003 model in a zero-shot scenario on WNUT-17 where we find that it outperforms GPT-3.5 and achieves comparable performance to GPT-4. In a zero-shot setting, ELLEN also achieves over 75% of the performance of a strong, fully supervised model trained on gold data. Our code is available at: https://github.com/hriaz17/ELLEN.

This advancement in natural language processing technology has significantly improved the functionality of conversational systems. Focusing more on language model development has spurred significant progress in understanding and generating human language, pushing the boundaries of machine interaction. This breakthrough primarily involves the enhancement of text-based applications, demonstrating how AI can more effectively comprehend, interpret, and respond to human language. This shift has had a profound impact on various sectors, ranging from customer service automation to sophisticated data analysis, emphasizing the transformative power of advanced language models in AI. Due to the success of ChatGPT, the training strategy of fine-tuning with instructions[2] has also been endorsed by researchers in the language model domain.

We empirically study a simple layer-pruning strategy for popular families of open-weight pretrained LLMs, finding minimal degradation of performance on different question-answering benchmarks until after a large fraction (up to half) of the layers are removed. To prune these models, we identify the optimal block of layers to prune by considering similarity across layers; then, to "heal" the damage, we perform a small amount of finetuning. In particular, we use parameter-efficient finetuning (PEFT) methods, specifically quantization and Low Rank Adapters (QLoRA), such that each of our experiments can be performed on a single A100 GPU. From a practical perspective, these results suggest that layer pruning methods can complement other PEFT strategies to further reduce computational resources of finetuning on the one hand, and can improve the memory and latency of inference on the other hand. From a scientific perspective, the robustness of these LLMs to the deletion of layers implies either that current pretraining methods are not properly leveraging the parameters in the deeper layers of the network or that the shallow layers play a critical role in storing knowledge.

The English-speaking world was rocked by the advent of ChatGPT in November 2022. Here, suddenly, was a chatbot that could do a credible imitation of a human being, producing text that seemed like it was written by a real person. People worried that ChatGPT, Google's Bard, and the like would cause widespread cheating as students turned over their writing assignments to a machine, or that they would lead to the mass production of misinformation and propaganda, outstripping the abilities of Russian troll farms. Those concerns arise in languages other than English as well. So far, however, chatbots based on Large Language Models (LLMs) appear to perform best in English, while sometimes struggling to mimic humans in other tongues.

Generative AI has been predicted to add trillions to the world economy in a productivity boost never before seen in history (if it doesn't wipe out humanity first). A growing number of sceptics, including some leading AI scientists, are wondering whether the tech might not deliver on its promises to boost the world economy. Goldman Sachs famously predicted that generative AI would bring about'sweeping changes' to the world economy, driving a 7 trillion increase in global GDP and lifting productivity growth by 1.5 percent this decade. Professor Gary Marcus of New York University wrote on Substack that'we are starting to see signs' that generative AI might be a'dud'. Among the warning signs was a report in the Wall Street Journal suggesting that customers found the 30 a month price of Microsoft's new AI-boosted Copilot software too expensive.

Suddenly, "open source" is the latest buzzword in AI circles. Meta has pledged to create open-source artificial general intelligence. And Elon Musk is suing OpenAI over its lack of open-source AI models. Meanwhile, a growing number of tech leaders and companies are setting themselves up as open-source champions. But there's a fundamental problem--no one can agree on what "open-source AI" means.

Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human judgement, revealing that existing calibration methods aimed at mitigating biases are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PairS), an uncertainty-guided search method that employs LLMs to conduct pairwise comparisons and efficiently ranks candidate texts. PairS achieves state-of-the-art performance on representative evaluation tasks and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PairS benefits from calibration.

Automated program repair has emerged as a powerful technique to mitigate the impact of software bugs on system reliability and user experience. This paper introduces RepairAgent, the first work to address the program repair challenge through an autonomous agent based on a large language model (LLM). Unlike existing deep learning-based approaches, which prompt a model with a fixed prompt or in a fixed feedback loop, our work treats the LLM as an agent capable of autonomously planning and executing actions to fix bugs by invoking suitable tools. RepairAgent freely interleaves gathering information about the bug, gathering repair ingredients, and validating fixes, while deciding which tools to invoke based on the gathered information and feedback from previous fix attempts. Key contributions that enable RepairAgent include a set of tools that are useful for program repair, a dynamically updated prompt format that allows the LLM to interact with these tools, and a finite state machine that guides the agent in invoking the tools. Our evaluation on the popular Defects4J dataset demonstrates RepairAgent's effectiveness in autonomously repairing 164 bugs, including 39 bugs not fixed by prior techniques. Interacting with the LLM imposes an average cost of 270,000 tokens per bug, which, under the current pricing of OpenAI's GPT-3.5 model, translates to 14 cents of USD per bug. To the best of our knowledge, this work is the first to present an autonomous, LLM-based agent for program repair, paving the way for future agent-based techniques in software engineering.

Large language models (LLMs) have become a significant workload since their appearance. However, they are also computationally expensive as they have billions of parameters and are trained with massive amounts of data. Thus, recent works have developed dedicated CUDA kernels for LLM training and inference instead of relying on compilergenerated ones, so that hardware resources are as fully utilized as possible. In this work, we explore the possibility of GPU native instruction optimization to further push the CUDA kernels to extreme performance. Contrary to prior works, we adopt an automatic optimization approach by defining a search space of possible GPU native instruction schedules, and then we apply stochastic search to perform optimization. Experiments show that SIP can further improve CUDA kernel throughput by automatically discovering better GPU native instruction schedules and the optimized schedules are tested by 10 million test samples.

Aspect-Based Sentiment Analysis (ABSA) aims to identify terms or multiword expressions (MWEs) on which sentiments are expressed and the sentiment polarities associated with them. The development of supervised models has been at the forefront of research in this area. However, training these models requires the availability of manually annotated datasets which is both expensive and time-consuming. Furthermore, the available annotated datasets are tailored to a specific domain, language, and text type. In this work, we address this notable challenge in current state-of-the-art ABSA research. We propose a hybrid approach for Aspect Based Sentiment Analysis using transfer learning. The approach focuses on generating weakly-supervised annotations by exploiting the strengths of both large language models (LLM) and traditional syntactic dependencies. We utilise syntactic dependency structures of sentences to complement the annotations generated by LLMs, as they may overlook domain-specific aspect terms. Extensive experimentation on multiple datasets is performed to demonstrate the efficacy of our hybrid method for the tasks of aspect term extraction and aspect sentiment classification.