Generative artificial intelligence (Gen AI) systems represent a critical technology with far-reaching implications across multiple domains of society. However, their deployment entails a range of risks and challenges that require careful evaluation. To date, there has been a lack of comprehensive, interdisciplinary studies offering a systematic comparison between open-source and proprietary (closed) generative AI systems, particularly regarding their respective advantages and drawbacks. This study aims to: i) critically evaluate and compare the characteristics, opportunities, and challenges of open and closed generative AI models; and ii) propose foundational elements for the development of an Open, Public, and Safe Gen AI framework. As a methodology, we adopted a combined approach that integrates three methods: literature review, critical analysis, and comparative analysis. The proposed framework outlines key dimensions, openness, public governance, and security, as essential pillars for shaping the future of trustworthy and inclusive Gen AI. Our findings reveal that open models offer greater transparency, auditability, and flexibility, enabling independent scrutiny and bias mitigation. In contrast, closed systems often provide better technical support and ease of implementation, but at the cost of unequal access, accountability, and ethical oversight. The research also highlights the importance of multi-stakeholder governance, environmental sustainability, and regulatory frameworks in ensuring responsible development.

Agents are predominantly evaluated and optimized via task success metrics, which are coarse, rely on manual design from experts, and fail to reward intermediate emergent behaviors. We propose **AutoLibra**, a framework for agent evaluation, that transforms open-ended human feedback *e.g.* "If you find that the button is disabled, don't click it again", or "This agent has too much autonomy to decide what to do on its own" into metrics for evaluating fine-grained behaviors in agent trajectories. AutoLibra accomplishes this by grounding feedback to an agent's behavior, clustering similar positive and negative behaviors, and creating concrete metrics with clear definitions and concrete examples, which can be used for prompting LLM-as-a-Judge as evaluators. We further propose two meta metrics to evaluate the alignment of a set of (induced) metrics with open feedback: "coverage" and "redundancy". Through optimizing these meta-metrics, we experimentally demonstrate AutoLibra's ability to induce more concrete agent evaluation metrics than the ones proposed in previous agent evaluation benchmarks and discover new metrics to analyze agents. We also present two applications of AutoLibra in agent improvement: First, we show that AutoLibra serve human prompt engineers for diagonalize agent failures and improve prompts iterative. Moreover, we find that AutoLibra can induce metrics for automatic optimization for agents, which makes agents improve through self-regulation. Our results suggest that AutoLibra is a powerful task-agnostic tool for evaluating and improving language agents.

To what extent can entire books be extracted from LLMs? Using the Llama 3 70B family of models, and the "prefix-prompting" extraction technique, we were able to auto-regressively reconstruct, with a very high level of similarity, one entire book (Alice's Adventures in Wonderland) from just the first 500 tokens. We were also able to obtain high extraction rates on several other books, piece-wise. However, these successes do not extend uniformly to all books. We show that extraction rates of books correlate with book popularity and thus, likely duplication in the training data. We also confirm the undoing of mitigations in the instruction-tuned Llama 3.1, following recent work (Nasr et al., 2025). We further find that this undoing comes from changes to only a tiny fraction of weights concentrated primarily in the lower transformer blocks. Our results provide evidence of the limits of current regurgitation mitigation strategies and introduce a framework for studying how fine-tuning affects the retrieval of verbatim memorization in aligned LLMs.

Recently, Large Language Models (LLMs) have dominated much of the artificial intelligence scene with their ability to process and generate natural languages. However, the majority of LLM research and development remains English-centric, leaving low-resource languages such as those in the Southeast Asian (SEA) region under-represented. To address this representation gap, we introduce Llama-SEA-LION-v3-8B-IT and Gemma-SEA-LION-v3-9B-IT, two cutting-edge multilingual LLMs designed for SEA languages. The SEA-LION family of LLMs supports 11 SEA languages, namely English, Chinese, Indonesian, Vietnamese, Malay, Thai, Burmese, Lao, Filipino, Tamil, and Khmer. Our work leverages large-scale multilingual continued pre-training with a comprehensive post-training regime involving multiple stages of instruction fine-tuning, alignment, and model merging. Evaluation results on multilingual benchmarks indicate that our models achieve state-of-the-art performance across LLMs supporting SEA languages. We open-source the models to benefit the wider SEA community.

Pulsar surveys generate millions of candidates per run, overwhelming manual inspection. This thesis builds a deep learning pipeline for radio pulsar candidate selection that fuses array-derived features with image diagnostics. From approximately 500 GB of Giant Metrewave Radio Telescope (GMRT) data, raw voltages are converted to filterbanks (SIGPROC), then de-dispersed and folded across trial dispersion measures (PRESTO) to produce approximately 32,000 candidates. Each candidate yields four diagnostics--summed profile, time vs. phase, subbands vs. phase, and DM curve--represented as arrays and images. A baseline stacked model (ANNs for arrays + CNNs for images with logistic-regression fusion) reaches 68% accuracy. We then refine the CNN architecture and training (regularization, learning-rate scheduling, max-norm constraints) and mitigate class imbalance via targeted augmentation, including a GAN-based generator for the minority class. The enhanced CNN attains 87% accuracy; the final GAN+CNN system achieves 94% accuracy with balanced precision and recall on a held-out test set, while remaining lightweight enough for near--real-time triage. The results show that combining array and image channels improves separability over image-only approaches, and that modest generative augmentation substantially boosts minority (pulsar) recall. The methods are survey-agnostic and extensible to forthcoming high-throughput facilities.

Determining the precise geographic location of an image at a global scale remains an unsolved challenge. Standard image retrieval techniques are inefficient due to the sheer volume of images (>100M) and fail when coverage is insufficient. Scalable solutions, however, involve a trade-off: global classification typically yields coarse results (10+ kilometers), while cross-view retrieval between ground and aerial imagery suffers from a domain gap and has been primarily studied on smaller regions. This paper introduces a hybrid approach that achieves fine-grained geo-localization across a large geographic expanse the size of a continent. We leverage a proxy classification task during training to learn rich feature representations that implicitly encode precise location information. We combine these learned prototypes with embeddings of aerial imagery to increase robustness to the sparsity of ground-level data. This enables direct, fine-grained retrieval over areas spanning multiple countries. Our extensive evaluation demonstrates that our approach can localize within 200m more than 68\% of queries of a dataset covering a large part of Europe. The code is publicly available at https://scaling-geoloc.github.io.

We study the ability of Transformer models to learn sequences generated by Permuted Congruential Generators (PCGs), a widely used family of pseudo-random number generators (PRNGs). PCGs introduce substantial additional difficulty over linear congruential generators (LCGs) by applying a series of bit-wise shifts, XORs, rotations and truncations to the hidden state. We show that Transformers can nevertheless successfully perform in-context prediction on unseen sequences from diverse PCG variants, in tasks that are beyond published classical attacks. Surprisingly, we find even when the output is truncated to a single bit, it can be reliably predicted by the model. When multiple distinct PRNGs are presented together during training, the model can jointly learn them, identifying structures from different permutations. We demonstrate a scaling law with modulus m: the number of in-context sequence elements required for near-perfect prediction grows as m. Finally, we analyze embedding layers and uncover a novel clustering phenomenon: the model spontaneously groups the integer inputs into bitwise rotationally-invariant clusters, revealing how representations can transfer from smaller to larger moduli. Transformer-based models have achieved remarkable success across language, vision, and algorithmic tasks, demonstrating an ability to capture complex patterns from large-scale data (V aswani et al., 2023; Dosovitskiy et al., 2021). Beyond supervised training, they can acquire new behaviors directly from examples provided in the input, a phenomenon known as in-context learning (Brown et al., 2020; Olsson et al., 2022). Despite these successes, fundamental questions remain: what kinds of patterns can Transformers reliably learn, how can we train them efficiently and what mechanisms underlie their ability to generalize? To address these questions, we use pseudo-random number generators (PRNGs) as a controlled benchmark.

Reinforcement learning (RL) fine-tuning of large language models (LLMs) often suffers from instability due to the numerical mismatch between the training and inference policies. While prior work has attempted to mitigate this issue through algorithmic corrections or engineering alignments, we show that its root cause lies in the floating point precision itself. The widely adopted BF16, despite its large dynamic range, introduces large rounding errors that breaks the consistency between training and inference. In this work, we demonstrate that simply reverting to \textbf{FP16} effectively eliminates this mismatch. The change is simple, fully supported by modern frameworks with only a few lines of code change, and requires no modification to the model architecture or learning algorithm. Our results suggest that using FP16 uniformly yields more stable optimization, faster convergence, and stronger performance across diverse tasks, algorithms and frameworks. We hope these findings motivate a broader reconsideration of precision trade-offs in RL fine-tuning.

AIs have made rapid progress on research-oriented benchmarks of knowledge and reasoning, but it remains unclear how these gains translate into economic value and automation. To measure this, we introduce the Remote Labor Index (RLI), a broadly multi-sector benchmark comprising real-world, economically valuable projects designed to evaluate end-to-end agent performance in practical settings. AI agents perform near the floor on RLI, with the highest-performing agent achieving an automation rate of 2.5%. These results help ground discussions of AI automation in empirical evidence, setting a common basis for tracking AI impacts and enabling stakeholders to proactively navigate AI-driven labor automation.

Age-related macular degeneration (AMD) is one of the leading causes of irreversible vision impairment in people over the age of 60. This research focuses on semantic segmentation for AMD lesion detection in RGB fundus images, a non-invasive and cost-effective imaging technique. The results of the ADAM challenge - the most comprehensive AMD detection from RGB fundus images research competition and open dataset to date - serve as a benchmark for our evaluation. Taking the U-Net connectivity as a base of our framework, we evaluate and compare several approaches to improve the segmentation model's architecture and training pipeline, including pre-processing techniques, encoder (backbone) deep network types of varying complexity, and specialized loss functions to mitigate class imbalances on image and pixel levels. The main outcome of this research is the final configuration of the AMD detection framework, which outperforms all the prior ADAM challenge submissions on the multi-class segmentation of different AMD lesion types in non-invasive RGB fundus images. The source code used to conduct the experiments presented in this paper is made freely available.