-
Peeking Behind Closed Doors: Risks of LLM Evaluation by Private Data Curators
A critical examination of the risks and challenges posed by private evaluators (for example ScaleAI) in the LLM landscape, highlighting financial incentives, conflicts of interest, and prevalence of evaluation biases even when acting in good faith.
-
Pitfalls of Evidence-Based AI Policy
Evidence is of irreplaceable value to policymaking. However, there are systematic biases shaping the evidence that the AI community produces. Holding regulation to too high an evidentiary standard can lead to systmatic neglect of certain risks. If the goal is evidence-based AI policy, the first regulatory objective must be to actively facilitate the process of identifying, studying, and deliberating about AI risks.
-
Positional Embeddings in Transformer Models: Evolution from Text to Vision Domains
Positional encoding has become an essential element in transformer models, addressing their fundamental property of permutation invariance and allowing them to understand sequential relationships within data. This blog post examines positional encoding techniques, emphasizing their vital importance in traditional transformers and their use with 2D data in Vision Transformers (ViT). We explore two contemporary methods—ALiBi (Attention with Linear Biases) and RoPE (Rotary Position Embedding)—analyzing their unique approaches to tackling the challenge of sequence length extrapolation during inference, a significant issue for transformers. Additionally, we compare these methods' fundamental similarities and differences, assessing their impact on transformer performance across various fields. We also look into how interpolation strategies have been utilized to enhance the extrapolation capabilities of these methods; we conclude this blog with an empirical comparison of ALiBi and RoPE in Vision Transformers. To the best of our knowledge, this represents the first direct comparison of these positional encoding methods with those used in standard Vision Transformers.
-
Pre-training of Foundation Adapters for LLM Fine-tuning
Adapter-based fine-tuning methods insert small, trainable adapters into frozen pre-trained LLMs, significantly reducing computational costs while maintaining performance. However, despite these advantages, traditional adapter fine-tuning suffers from training instability due to random weight initialization. This instability can lead to inconsistent performance across different runs. Therefore, to address this issue, this blog post introduces pre-trained foundation adapters as a technique for weight initialization. This technique potentially improves the efficiency and effectiveness of the fine-tuning process. Specifically, we combine continual pre-training and knowledge distillation to pre-train foundation adapters. Experiments confirm the effectiveness of this approach across multiple tasks. Moreover, we highlight the advantage of using pre-trained foundation adapter weights over random initialization specifically in a summarization task.
-
Reassessing EMNLP 2024’s Best Paper: Does Divergence-Based Calibration for Membership Inference Attacks Hold Up?
TL;DR: No.
A critical analysis of the EMNLP Best Paper proposing a divergence-based calibration for Membership Inference Attacks (MIAs). We explore its experimental shortcomings, issues with temporally shifted benchmarks, and what this means for machine learning awards. -
Reexamining the Aleatoric and Epistemic Uncertainty Dichotomy
When discussing uncertainty estimates for the safe deployment of AI agents in the real world, the field typically distinguishes between aleatoric and epistemic uncertainty. This dichotomy may seem intuitive and well-defined at first glance, but this blog post reviews examples, quantitative findings, and theoretical arguments that reveal that popular definitions of aleatoric and epistemic uncertainties directly contradict each other and are intertwined in fine nuances. We peek beyond the epistemic and aleatoric uncertainty dichotomy and reveal a spectrum of uncertainties that help solve practical tasks especially in the age of large language models.
-
Repurposing in AI: A Distinct Approach or an Extension of Creative Problem Solving?
Creativity is defined as the ability to produce novel, useful, and surprising ideas. A sub area of creativity is creative problem solving, the capacity of an agent to discover novel and previously unseen ways to accomplish a task, according to its perspective. However, there is a related concept, repurposing, that has often been overlooked in the broader context of creative problem solving in AI. Repurposing involves identifying and utilizing existing objects, resources, or processes in innovative ways to address different problems. While these two concepts may seem distinct at first glance, recent studies in creativity in AI suggest that they may be more closely intertwined than previously thought. By examining the underlying mechanisms and cognitive processes involved in both creative problem solving and repurposing, we can begin to understand how these approaches complement each other.
-
Restating the Proof of Linear Convergence for Linear GNNs
We lead the readers through the core proof of a pioneering paper that studies the training dynamics of linear GNNs. First, we reorganize the proof and provide a more concise and reader-friendly version, highlighting several key components. In doing so, we identify a hidden error and correct it, demonstrating that it has no impact on the main result. Additionally, we offer a dialectical discussion on the strengths and an overlooked aspect of the approach.
-
Rethinking Graph Prompts: Unraveling the Power of Data Manipulation in Graph Neural Networks
Graph Neural Networks (GNNs) have transformed graph learning but face challenges like distribution shifts, data anomalies, and adversarial vulnerabilities. Graph prompt emerges as a novel solution, enabling data transformation to align graph data with pre-trained models without altering model parameters. This paradigm addresses negative transfer, enhances adaptability, and bridges modality gaps. Unlike traditional fine-tuning, graph prompts rewrite graph structures and features through components like prompt tokens and insertion patterns, improving flexibility and efficiency. Applications in IoT, drug discovery, fraud detection, and personalized learning demonstrate their potential to dynamically adapt graph data. While promising, challenges such as optimal design, benchmarks, and gradient issues persist. Addressing these will unlock full potential of graph prompt to advance GNNs for complex real-world tasks.
-
Sample Blog Post
Your blog post's abstract. Please add your abstract or summary here and not in the main body of your text. Do not include math/latex or hyperlinks.
-
Sample Blog Post (HTML version)
Your blog post's abstract. Please add your abstract or summary here and not in the main body of your text. Do not include math/latex or hyperlinks.
-
SPD Attack - Prevention of AI Powered Image Editing by Image Immunization
Recent advances in image-to-image editing models offer both benefits and risks. While they enhance creativity, accessibility, and applications in fields ranging from medicine to environmental science, they can also enable misuse, such as identity manipulation, copyright infringement, and deepfake creation. This blog explores methods to protect images from such misuse, reproduces findings from relevant research, and extends them across various models and datasets.