🥇Top ML Papers of the Week
The top ML Papers of the Week (Sep 25 - Oct 1)
1). The Reversal Curse - finds that LLMs trained on sentences of the form “A is B” will not automatically generalize to the reverse direction “B is A”, i.e., the Reversal Curse; shows the effect through finetuning LLMs on fictitious statements and demonstrating its robustness across model sizes and model families. (paper | tweet)
2). Effective Long-Context Scaling with LLMs - propose a 70B variant that can already surpass gpt-3.5-turbo-16k’s overall performance on a suite of long-context tasks. This involves a cost-effective instruction tuning procedure that does not require human-annotated long instruction data. (paper | tweet)
3). Graph Neural Prompting with LLMs - proposes a plug-and-play method to assist pre-trained LLMs in learning beneficial knowledge from knowledge graphs (KGs); includes various designs, including a standard graph neural network encoder, a cross-modality pooling module, a domain projector, and a self-supervised link prediction objective. (paper | tweet)
4). Vision Transformers Need Registers - identifies artifacts in feature maps of vision transformer networks that are repurposed for internal computations; this work proposes a solution to provide additional tokens to the input sequence to fill that role; the solution fixes the problem, leads to smoother feature and attention maps, and sets new state-of-the-art results on dense visual prediction tasks. (paper | tweet)
5). Boolformer - presents the first Transformer architecture trained to perform end-to-end symbolic regression of Boolean functions; it can predict compact formulas for complex functions and be applied to modeling the dynamics of gene regulatory networks. (paper | tweet)
Reach out to hello@dair.ai if you would like to sponsor the next issue of the newsletter. We can help promote your AI tool, research, or company to ~10K AI researchers and practitioners.
6). LlaVA-RLHF - adapts factually augmented RLHF to aligning large multimodal models; this approach alleviates the reward hacking in RLHF and improves performance on the LlaVA-Bench dataset with the 94% performance level of the text-only GPT-4. (paper | tweet)
7). LLM Alignment Survey - a comprehensive survey paper on LLM alignment; topics include Outer Alignment, Inner Alignment, Mechanistic Interpretability, Attacks on Aligned LLMs, Alignment Evaluation, Future Directions, and Discussions. (paper | tweet)
8). Qwen LLM - proposes a series of LLMs demonstrating the strength of RLHF on tasks involving tool use and planning capabilities for creating language agents. (paper | tweet)
9). MentalLlaMa - an open-source LLM series for interpretable mental health analysis with instruction-following capability; it also proposes a multi-task and multi-source interpretable mental health instruction dataset on social media with 105K data samples. (paper | tweet)
10). Logical Chain-of-Thought in LLMs - a new neurosymbolic framework to improve zero-shot chain-of-thought reasoning in LLMs; leverages principles from symbolic logic to verify and revise reasoning processes to improve the reasoning capabilities of LLMs. (paper | tweet)