Source code comes in different shapes and forms. Previous research has already shown code to be more predictable than natural language as well as highlighted its statistical predictability at the token level: source code can be natural. More recently, the structure of code -- control flow, syntax graphs, abstract syntax trees etc. -- has been successfully used to improve the state-of-the-art on numerous tasks: code suggestion, code summarisation, method naming etc. This body of work implicitly assumes that structured representations of code are similarly statistically predictable, i.e. that a structured view of code is also natural. We consider that this view should be made explicit and propose directly studying the Structured Naturalness Hypothesis. Beyond just naming existing research that assumes this hypothesis and formulating it, we also provide evidence in the case of trees: TreeLSTM models over ASTs for some languages, such as Ruby, are competitive with $n$-gram models while handling the syntax token issue highlighted by previous research 'for free'. For other languages, such as Java or Python, we find tree models to perform worse, suggesting that downstream task improvement is uncorrelated to the language modelling task. Further, we show how such naturalness signals can be employed for near state-of-the-art results on just-in-time defect prediction while forgoing manual feature engineering work.

This study introduces a framework that leverages AI-generated descriptive codes to indicate a text's fecundity--the density of unique human-generated codes--in thematic analysis. Rather than replacing human interpretation, AI-generated codes guide the selection of texts likely to yield richer qualitative insights. Using a dataset of 2,530 Malaysian news articles on refugee attitudes, we compare AI-selected documents to randomly chosen ones by having three human coders independently derive codes. The results demonstrate that AI-selected texts exhibit approximately twice the fecundity. Our findings support the use of AI-generated codes as an effective proxy for identifying documents with a high potential for meaning-making in thematic analysis.

In this paper, we conducted a Multi-Perspective Comparative Narrative Analysis (CNA) on three prominent LLMs: GPT-3.5, PaLM2, and Llama2. We applied identical prompts and evaluated their outputs on specific tasks, ensuring an equitable and unbiased comparison between various LLMs. Our study revealed that the three LLMs generated divergent responses to the same prompt, indicating notable discrepancies in their ability to comprehend and analyze the given task. Human evaluation was used as the gold standard, evaluating four perspectives to analyze differences in LLM performance.

How Good Are Large Language Models for Course Recommendation in MOOCs? Shin'ichi Konomi Kyushu University, Japan konomi@artsci.kyushu-u.ac.jp ABSTRACT Large Language Models (LLMs) have made significant strides in natural language processing and are increasingly being integrated into recommendation systems. However, their potential in educational recommendation systems has yet to be fully explored. This paper investigates the use of LLMs as a general-purpose recommendation model, leveraging their vast knowledge derived from large-scale corpora for course recommendation tasks. We explore a variety of approaches, ranging from prompt-based methods to more advanced fine-tuning techniques, and compare their performance against traditional recommendation models. Extensive experiments were conducted on a real-world MOOC dataset, evaluating using LLMs as course recommendation systems across key dimensions such as accuracy, diversity, and novelty. Our results demonstrate that LLMs can achieve good performance comparable to traditional models, highlighting their potential to enhance educational recommendation systems.

Architectural Knowledge Management (AKM) is crucial for software development but remains challenging due to the lack of standardization and high manual effort. Architecture Decision Records (ADRs) provide a structured approach to capture Architecture Design Decisions (ADDs), but their adoption is limited due to the manual effort involved and insufficient tool support. Our previous work has shown that Large Language Models (LLMs) can assist in generating ADDs. However, simply prompting the LLM does not produce quality ADDs. Moreover, using third-party LLMs raises privacy concerns, while self-hosting them poses resource challenges. To this end, we experimented with different approaches like few-shot, retrieval-augmented generation (RAG) and fine-tuning to enhance LLM's ability to generate ADDs. Our results show that both techniques improve effectiveness. Building on this, we propose Domain Specific Retreival Augumented Few Shot Fine Tuninng, DRAFT, which combines the strengths of all these three approaches for more effective ADD generation. DRAFT operates in two phases: an offline phase that fine-tunes an LLM on generating ADDs augmented with retrieved examples and an online phase that generates ADDs by leveraging retrieved ADRs and the fine-tuned model. We evaluated DRAFT against existing approaches on a dataset of 4,911 ADRs and various LLMs and analyzed them using automated metrics and human evaluations. Results show DRAFT outperforms all other approaches in effectiveness while maintaining efficiency. Our findings indicate that DRAFT can aid architects in drafting ADDs while addressing privacy and resource constraints.

Recent advances in long-context models (LCMs), designed to handle extremely long input contexts, primarily focus on utilizing external contextual information, often leaving the influence of large language models' intrinsic knowledge underexplored. In this work, we investigate how this intrinsic knowledge affects content generation and demonstrate that its impact becomes increasingly pronounced as context length extends. Furthermore, we show that the model's ability to utilize intrinsic knowledge, which we call intrinsic retrieval ability, does not improve simultaneously with its ability to leverage contextual knowledge through extrinsic retrieval ability. Moreover, better extrinsic retrieval can interfere with the model's ability to use its own knowledge effectively, limiting its full potential. To bridge this gap, we design a simple yet effective Hybrid Needle-in-a-Haystack test that evaluates models based on their capabilities across both retrieval abilities, rather than solely emphasizing extrinsic retrieval ability. Our experimental results reveal that Qwen-2.5 models significantly outperform Llama-3.1 models, demonstrating superior intrinsic retrieval ability. Moreover, even the more powerful Llama-3.1-70B-Instruct model fails to exhibit better performance under LCM conditions, highlighting the importance of evaluating models from a dual-retrieval perspective.

Machine unlearning is a promising approach to improve LLM safety by removing unwanted knowledge from the model. However, prevailing gradient-based unlearning methods suffer from issues such as high computational costs, hyperparameter instability, poor sequential unlearning capability, vulnerability to relearning attacks, low data efficiency, and lack of interpretability. While Sparse Autoencoders are well-suited to improve these aspects by enabling targeted activation-based unlearning, prior approaches underperform gradient-based methods. This work demonstrates that, contrary to these earlier findings, SAEs can significantly improve unlearning when employed dynamically. We introduce $\textbf{Dynamic DAE Guardrails}$ (DSG), a novel method for precision unlearning that leverages principled feature selection and a dynamic classifier. Our experiments show DSG substantially outperforms leading unlearning methods, achieving superior forget-utility trade-offs. DSG addresses key drawbacks of gradient-based approaches for unlearning -- offering enhanced computational efficiency and stability, robust performance in sequential unlearning, stronger resistance to relearning attacks, better data efficiency including zero-shot settings, and more interpretable unlearning.

SYNTHFM: TRAINING MODALITY -AGNOSTIC FOUNDA TION MODELS FOR MEDICAL IMAGE SEGMENT A TION WITHOUT REAL MEDICAL DA T A Sourya Sengupta 1, 2, Satrajit Chakrabarty 1, Keerthi Sravan Ravi 1, Gopal Avinash 1, Ravi Soni 1 1 GE HealthCare, San Ramon, CA, USA 2 University of Illinois Urbana-Champaign, Urbana, IL, USA ABSTRACT Foundation models like the Segment Anything Model (SAM) excel in zero-shot segmentation for natural images but struggle with medical image segmentation due to differences in texture, contrast, and noise. Annotating medical images is costly and requires domain expertise, limiting large-scale annotated data availability. To address this, we propose Syn-thFM, a synthetic data generation framework that mimics the complexities of medical images, enabling foundation models to adapt without real medical data. Using SAM's pretrained encoder and training the decoder from scratch on SynthFM's dataset, we evaluated our method on 11 anatomical structures across 9 datasets (CT, MRI, and Ultrasound). SynthFM outperformed zero-shot baselines like SAM and MedSAM, achieving superior results under different prompt settings and on out-of-distribution datasets.

Children can acquire language from less than 100 million words of input. Large language models are far less data-efficient: they typically require 3 or 4 orders of magnitude more data and still do not perform as well as humans on many evaluations. These intensive resource demands limit the ability of researchers to train new models and use existing models as developmentally plausible cognitive models. The BabyLM Challenge is a communal effort in which participants compete to optimize language model training on a fixed data budget. Submissions are compared on various evaluation tasks targeting grammatical ability, downstream task performance, and generalization. Participants can submit to up to three tracks with progressively looser data restrictions. From over 30 submissions, we extract concrete recommendations on how best to train data-efficient language models, and on where future efforts should (and perhaps should not) focus. The winning submissions using the LTG-BERT architecture (Samuel et al., 2023) outperformed models trained on trillions of words. Other submissions achieved strong results through training on shorter input sequences or training a student model on a pretrained teacher. Curriculum learning attempts, which accounted for a large number of submissions, were largely unsuccessful, though some showed modest improvements.

Large language models (LLMs) have gained significant interest in industry due to their impressive capabilities across a wide range of tasks. However, the widespread adoption of LLMs presents several challenges, such as integration into existing applications and infrastructure, utilization of company proprietary data, models, and APIs, and meeting cost, quality, responsiveness, and other requirements. To address these challenges, there is a notable shift from monolithic models to compound AI systems, with the premise of more powerful, versatile, and reliable applications. However, progress thus far has been piecemeal, with proposals for agentic workflows, programming models, and extended LLM capabilities, without a clear vision of an overall architecture. In this paper, we propose a 'blueprint architecture' for compound AI systems for orchestrating agents and data for enterprise applications. In our proposed architecture the key orchestration concept is 'streams' to coordinate the flow of data and instructions among agents. Existing proprietary models and APIs in the enterprise are mapped to 'agents', defined in an 'agent registry' that serves agent metadata and learned representations for search and planning. Agents can utilize proprietary data through a 'data registry' that similarly registers enterprise data of various modalities. Tying it all together, data and task 'planners' break down, map, and optimize tasks and queries for given quality of service (QoS) requirements such as cost, accuracy, and latency. We illustrate an implementation of the architecture for a use-case in the HR domain and discuss opportunities and challenges for 'agentic AI' in the enterprise.