Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones trained on more tokens. What accounts for this? To quantify the impact of these design choices, we meta-analyze 92 open-source pretrained models across a wide array of scales, including state-of-the-art open-weights models as well as less performant models and those with less conventional design decisions. We find that by incorporating features besides model size and number of training tokens, we can achieve a relative 3-28% increase in ability to predict downstream performance compared with using scale alone. Analysis of model design decisions reveal insights into data composition, such as the trade-off between language and code tasks at 15-25\% code, as well as the better performance of some architectural decisions such as choosing rotary over learned embeddings. Broadly, our framework lays a foundation for more systematic investigation of how model development choices shape final capabilities.

Researchers and practitioners in natural language processing and computational linguistics frequently observe and analyze the real language usage in large-scale corpora. For that purpose, they often employ off-the-shelf pattern-matching tools, such as grep, and keyword-in-context concordancers, which is widely used in corpus linguistics for gathering examples. Nonetheless, these existing techniques rely on surface-level string matching, and thus they suffer from the major limitation of not being able to handle orthographic variations and paraphrasing -- notable and common phenomena in any natural language. In addition, existing continuous approaches such as dense vector search tend to be overly coarse, often retrieving texts that are unrelated but share similar topics. Given these challenges, we propose a novel algorithm that achieves \emph{soft} (or semantic) yet efficient pattern matching by relaxing a surface-level matching with word embeddings. Our algorithm is highly scalable with respect to the size of the corpus text utilizing inverted indexes. We have prepared an efficient implementation, and we provide an accessible web tool. Our experiments demonstrate that the proposed method (i) can execute searches on billion-scale corpora in less than a second, which is comparable in speed to surface-level string matching and dense vector search; (ii) can extract harmful instances that semantically match queries from a large set of English and Japanese Wikipedia articles; and (iii) can be effectively applied to corpus-linguistic analyses of Latin, a language with highly diverse inflections.

Current approaches to emotion detection often overlook the inherent subjectivity of affective experiences, instead relying on aggregated labels that mask individual variations in emotional responses. We introduce iNews, a novel large-scale dataset explicitly capturing subjective affective responses to news headlines. Our dataset comprises annotations from 291 demographically diverse UK participants across 2,899 multimodal Facebook news posts from major UK outlets, with an average of 5.18 annotators per sample. For each post, annotators provide multifaceted labels including valence, arousal, dominance, discrete emotions, content relevance judgments, sharing likelihood, and modality importance ratings (text, image, or both). Furthermore, we collect comprehensive annotator persona information covering demographics, personality, media trust, and consumption patterns, which explain 15.2% of annotation variance - higher than existing NLP datasets. Incorporating this information yields a 7% accuracy gain in zero-shot prediction and remains beneficial even with 32-shot. iNews will enhance research in LLM personalization, subjectivity, affective computing, and individual-level behavior simulation.

Personalizing large language models (LLMs) is essential for delivering tailored interactions that improve user experience. Many existing personalization methods require fine-tuning LLMs for each user, rendering them prohibitively expensive for widespread adoption. Although retrieval-based approaches offer a more compute-efficient alternative, they still depend on large, high-quality datasets that are not consistently available for all users. To address this challenge, we propose CHAMELEON, a scalable and efficient personalization approach that uses (1) self-generated personal preference data and (2) representation editing to enable quick and cost-effective personalization. Our experiments on various tasks, including those from the LaMP personalization benchmark, show that CHAMELEON efficiently adapts models to personal preferences, improving instruction-tuned models and outperforms two personalization baselines by an average of 40% across two model architectures.

Academic citations are widely used for evaluating research and tracing knowledge flows. Such uses typically rely on raw citation counts and neglect variability in citation types. In particular, citations can vary in their fidelity as original knowledge from cited studies may be paraphrased, summarized, or reinterpreted, possibly wrongly, leading to variation in how much information changes from cited to citing paper. In this study, we introduce a computational pipeline to quantify citation fidelity at scale. Using full texts of papers, the pipeline identifies citations in citing papers and the corresponding claims in cited papers, and applies supervised models to measure fidelity at the sentence level. Analyzing a large-scale multi-disciplinary dataset of approximately 13 million citation sentence pairs, we find that citation fidelity is higher when authors cite papers that are 1) more recent and intellectually close, 2) more accessible, and 3) the first author has a lower H-index and the author team is medium-sized. Using a quasi-experiment, we establish the "telephone effect" - when citing papers have low fidelity to the original claim, future papers that cite the citing paper and the original have lower fidelity to the original. Our work reveals systematic differences in citation fidelity, underscoring the limitations of analyses that rely on citation quantity alone and the potential for distortion of evidence.

Recent advances in Large Language Models (LLMs) have raised interest in their formal reasoning capabilities, particularly in mathematics. While closed LLMs like GPT-4 perform well on mathematical benchmarks, e.g., GSM8K, it remains unclear whether small to medium-sized open LLMs can achieve similar performance, questioning their reliability. To close this gap, we propose a post-training approach leveraging a mixture of opinions (MoO) from weaker ancillary LLMs to enhance a (relatively) stronger LLM's reasoning. For that, each post-training sample is augmented with Chain-of-Thought (CoT) reasoning steps and answers from ancillary LLMs, enabling the main LLM to learn from diverse perspectives. We compare MoO with standard supervised fine-tuning (SFT), few-shot prompting, and the Mixture of Agents (MoA) method on mathematical reasoning benchmarks. Our results show that incorporating weaker LLMs' opinions improves mathematical reasoning by an average of 5%, highlighting the value of diverse perspectives in reasoning tasks.

Fans of the 1980's-era sitcom A Different World were initially thrilled to hear that the Cosby Show spinoff was making the jump to Netflix, with all six seasons presented in HD quality. Then they started watching, and noticed there was, indeed, something very different about how A Different World looks on streaming. The text on signs looks strangely garbled. Faces in the background appear squished, sometimes even a tad monstrous. The image as a whole looks like an animated watercolor painting. Everything looks… well, weird, bordering on grotesque.

Large language models (LLMs) have demonstrated remarkable proficiency in mainstream academic disciplines such as mathematics, physics, and computer science. However, human knowledge encompasses over 200 specialized disciplines, far exceeding the scope of existing benchmarks. The capabilities of LLMs in many of these specialized fields-particularly in light industry, agriculture, and service-oriented disciplines-remain inadequately evaluated. To address this gap, we present SuperGPQA, a comprehensive benchmark that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines. Our benchmark employs a novel Human-LLM collaborative filtering mechanism to eliminate trivial or ambiguous questions through iterative refinement based on both LLM responses and expert feedback. Our experimental results reveal significant room for improvement in the performance of current state-of-the-art LLMs across diverse knowledge domains (e.g., the reasoning-focused model DeepSeek-R1 achieved the highest accuracy of 61.82% on SuperGPQA), highlighting the considerable gap between current model capabilities and artificial general intelligence. Additionally, we present comprehensive insights from our management of a large-scale annotation process, involving over 80 expert annotators and an interactive Human-LLM collaborative system, offering valuable methodological guidance for future research initiatives of comparable scope.

This study developed an algorithm capable of detecting a reference line (a 0.2 mm thick piano wire) to accurately determine the position of an automated installation robot within an elevator shaft. A total of 3,245 images were collected from the experimental tower of H Company, the leading elevator manufacturer in South Korea, and the detection performance was evaluated using four experimental approaches (GCH, GSCH, GECH, FCH). During the initial image processing stage, Gaussian blurring, sharpening filter, embossing filter, and Fourier Transform were applied, followed by Canny Edge Detection and Hough Transform. Notably, the method was developed to accurately extract the reference line by averaging the x-coordinates of the lines detected through the Hough Transform. This approach enabled the detection of the 0.2 mm thick piano wire with high accuracy, even in the presence of noise and other interfering factors (e.g., concrete cracks inside the elevator shaft or safety bars for filming equipment). The experimental results showed that Experiment 4 (FCH), which utilized Fourier Transform in the preprocessing stage, achieved the highest detection rate for the LtoL, LtoR, and RtoL datasets. Experiment 2(GSCH), which applied Gaussian blurring and a sharpening filter, demonstrated superior detection performance on the RtoR dataset. This study proposes a reference line detection algorithm that enables precise position calculation and control of automated robots in elevator shaft installation. Moreover, the developed method shows potential for applicability even in confined working spaces. Future work aims to develop a line detection algorithm equipped with machine learning-based hyperparameter tuning capabilities.

Frontier AI models -- highly capable foundation models at the cutting edge of AI development -- may pose severe risks to public safety, human rights, economic stability, and societal value in the coming years. These risks could arise from deliberate adversarial misuse, system failures, unintended cascading effects, or simultaneous failures across multiple models. In response to such risks, at the AI Seoul Summit in May 2024, 16 global AI industry organizations signed the Frontier AI Safety Commitments, and 27 nations and the EU issued a declaration on their intent to define these thresholds. To fulfill these commitments, organizations must determine and disclose ``thresholds at which severe risks posed by a model or system, unless adequately mitigated, would be deemed intolerable.'' To assist in setting and operationalizing intolerable risk thresholds, we outline key principles and considerations; for example, to aim for ``good, not perfect'' thresholds in the face of limited data on rapidly advancing AI capabilities and consequently evolving risks. We also propose specific threshold recommendations, including some detailed case studies, for a subset of risks across eight risk categories: (1) Chemical, Biological, Radiological, and Nuclear (CBRN) Weapons, (2) Cyber Attacks, (3) Model Autonomy, (4) Persuasion and Manipulation, (5) Deception, (6) Toxicity, (7) Discrimination, and (8) Socioeconomic Disruption. Our goal is to serve as a starting point or supplementary resource for policymakers and industry leaders, encouraging proactive risk management that prioritizes preventing intolerable risks (ex ante) rather than merely mitigating them after they occur (ex post).