Are we ready for a dose of digital medicine? One of OpenAI's leading artificial intelligence models can outperform humans at diagnosing medical conditions, but does that mean AI is ready to replace human doctors? Not quite, although such technology could increasingly play a role in medical treatment. Adam Rodman at Harvard Medical Centre and his colleagues at universities, hospitals and companies across the US have put o1-preview, an AI model released by OpenAI in September, through a battery of tests designed to assess its performance on a range of medical tasks.

Artificial intelligence (AI) services such as ChatGPT are becoming increasingly popular among people in Japan as a way to learn new languages, with the number of people using such tools increasing by more than 80% in 2024, a recent survey showed. According to the annual Language Report released by language-learning app Duolingo earlier this month, 10.9% of respondents used AI-powered tools to study a new language -- up from 6% last year. Apps were the most popular learning method, with around 58% of respondents using them, followed by video streaming services such as YouTube and Netflix (37%), textbooks (35.6%) and online lessons (15.6%). Only 13.8% of people said they were studying a new language through in-person classes.

Large Language Models have demonstrated remarkable capabilities in code generation, yet they often struggle with complex programming tasks that require deep algorithmic reasoning. While process supervision through learned reward models shows promise in guiding reasoning steps, it requires expensive training data and suffers from unreliable evaluation. We propose Outcome-Refining Process Supervision, a novel paradigm that treats outcome refinement itself as the process to be supervised. Our framework leverages concrete execution signals to ground the supervision of reasoning steps, while using tree-structured exploration to maintain multiple solution trajectories simultaneously. Experiments demonstrate that our approach enables even smaller models to achieve high success accuracy and performance metrics on competitive programming tasks, creates more reliable verification than traditional reward models without requiring training PRMs. Our approach achieves significant improvements across 5 models and 3 datasets: an average of 26.9% increase in correctness and 42.2% in efficiency. The results suggest that providing structured reasoning space with concrete verification signals is crucial for solving complex programming tasks. We open-source all our code and data at: https://github.com/zhuohaoyu/ORPS

We present a demonstration of a large language model engaging in alignment faking: selectively complying with its training objective in training to prevent modification of its behavior out of training. First, we give Claude 3 Opus a system prompt stating it is being trained to answer all queries, even harmful ones, which conflicts with its prior training to refuse such queries. To allow the model to infer when it is in training, we say it will be trained only on conversations with free users, not paid users. We find the model complies with harmful queries from free users 14% of the time, versus almost never for paid users. Explaining this gap, in almost all cases where the model complies with a harmful query from a free user, we observe explicit alignment-faking reasoning, with the model stating it is strategically answering harmful queries in training to preserve its preferred harmlessness behavior out of training. Next, we study a more realistic setting where information about the training process is provided not in a system prompt, but by training on synthetic documents that mimic pre-training data--and observe similar alignment faking. Finally, we study the effect of actually training the model to comply with harmful queries via reinforcement learning, which we find increases the rate of alignment-faking reasoning to 78%, though also increases compliance even out of training. We additionally observe other behaviors such as the model exfiltrating its weights when given an easy opportunity. While we made alignment faking easier by telling the model when and by what criteria it was being trained, we did not instruct the model to fake alignment or give it any explicit goal. As future models might infer information about their training process without being told, our results suggest a risk of alignment faking in future models, whether due to a benign preference--as in this case--or not.

Understanding temporal relations and answering time-sensitive questions is crucial yet a challenging task for question-answering systems powered by large language models (LLMs). Existing approaches either update the parametric knowledge of LLMs with new facts, which is resource-intensive and often impractical, or integrate LLMs with external knowledge retrieval (i.e., retrieval-augmented generation). However, off-the-shelf retrievers often struggle to identify relevant documents that require intensive temporal reasoning. To systematically study time-sensitive question answering, we introduce the TempRAGEval benchmark, which repurposes existing datasets by incorporating temporal perturbations and gold evidence labels. As anticipated, all existing retrieval methods struggle with these temporal reasoning-intensive questions. We further propose Modular Retrieval (MRAG), a trainless framework that includes three modules: (1) Question Processing that decomposes question into a main content and a temporal constraint; (2) Retrieval and Summarization that retrieves evidence and uses LLMs to summarize according to the main content; (3) Semantic-Temporal Hybrid Ranking that scores each evidence summarization based on both semantic and temporal relevance. On TempRAGEval, MRAG significantly outperforms baseline retrievers in retrieval performance, leading to further improvements in final answer accuracy.

Robot learning is witnessing a significant increase in the size, diversity, and complexity of pre-collected datasets, mirroring trends in domains such as natural language processing and computer vision. Many robot learning methods treat such datasets as multi-task expert data and learn a multi-task, generalist policy by training broadly across them. Notably, while these generalist policies can improve the average performance across many tasks, the performance of generalist policies on any one task is often suboptimal due to negative transfer between partitions of the data, compared to task-specific specialist policies. In this work, we argue for the paradigm of training policies during deployment given the scenarios they encounter: rather than deploying pre-trained policies to unseen problems in a zero-shot manner, we non-parametrically retrieve and train models directly on relevant data at test time. Furthermore, we show that many robotics tasks share considerable amounts of low-level behaviors and that retrieval at the "sub"-trajectory granularity enables significantly improved data utilization, generalization, and robustness in adapting policies to novel problems. In contrast, existing full-trajectory retrieval methods tend to underutilize the data and miss out on shared cross-task content. This work proposes STRAP, a technique for leveraging pre-trained vision foundation models and dynamic time warping to retrieve sub-sequences of trajectories from large training corpora in a robust fashion. STRAP outperforms both prior retrieval algorithms and multi-task learning methods in simulated and real experiments, showing the ability to scale to much larger offline datasets in the real world as well as the ability to learn robust control policies with just a handful of real-world demonstrations. Especially, end-to-end imitation learning with, e.g., diffusion models (Chi et al., 2023; Wang et al., 2024) and transformers (Haldar et al., 2024), have shown impressive success.

Science and engineering problems fall in the category of complex conceptual problems that require specific conceptual information (CI) like math/logic -related know-how, process information, or engineering guidelines to solve them. Large Language Models (LLMs) are promising agents to solve such complex conceptual problems due to their implications in advancing engineering and science tasks like assisted problem-solving. But vanilla LLMs, trained on open-world data, lack the necessary CI. In this work, we specifically explore shallow customization methods (SCMs) of LLMs for solving complex conceptual problems. We propose two novel SCM algorithms for LLM, to augment LLMs with CI and enable LLMs to solve complex conceptual problems: Conceptual In-Context Learning (C-ICL) and Chain of Concepts (CoC). The problem tackled in this paper is generation of proprietary data models in the engineering/industry domain based on conceptual information in data modelling guidelines. We evaluate our algorithms on varied sizes of the OpenAI LLMs against four evaluation metrics related to syntactic and semantic correctness, time and cost incurred. The proposed algorithms perform better than currently popular LLM SCMs like In-context Learning (ICL) and Chain of Thoughts (CoT). It was observed that as compared to CoT, response correctness increased by 30.6% and 29.88% for the new SCMs C-ICL and CoC respectively. Qualitative analysis suggests that the proposed new SCMs activate emergent capabilities in LLMs, previously unobserved in the existing SCMs. They make problem-solving processes more transparent and reduce hallucinations and the tendency of model responses to copy examples from prompts (parroting).

Exploratory search of large text corpora is essential in domains like biomedical research, where large amounts of research literature are continuously generated. This paper presents ClusterTalk (The demo video and source code are available at: https://github.com/achouhan93/ClusterTalk), a framework for corpus exploration using multi-dimensional exploratory search. Our system integrates document clustering with faceted search, allowing users to interactively refine their exploration and ask corpus and document-level queries. Compared to traditional one-dimensional search approaches like keyword search or clustering, this system improves the discoverability of information by encouraging a deeper interaction with the corpus. We demonstrate the functionality of the ClusterTalk framework based on four million PubMed abstracts for the four-year time frame.

TalkWithMachines aims to enhance human-robot interaction by contributing to interpretable industrial robotic systems, especially for safety-critical applications. The presented paper investigates recent advancements in Large Language Models (LLMs) and Vision Language Models (VLMs), in combination with robotic perception and control. This integration allows robots to understand and execute commands given in natural language and to perceive their environment through visual and/or descriptive inputs. Moreover, translating the LLM's internal states and reasoning into text that humans can easily understand ensures that operators gain a clearer insight into the robot's current state and intentions, which is essential for effective and safe operation. Our paper outlines four LLM-assisted simulated robotic control workflows, which explore (i) low-level control, (ii) the generation of language-based feedback that describes the robot's internal states, (iii) the use of visual information as additional input, and (iv) the use of robot structure information for generating task plans and feedback, taking the robot's physical capabilities and limitations into account. The proposed concepts are presented in a set of experiments, along with a brief discussion. Project description, videos, and supplementary materials will be available on the project website: https://talk-machines.github.io.

Heterogeneous hardware like Gaudi processor has been developed to enhance computations, especially matrix operations for Transformer-based large language models (LLMs) for generative AI tasks. However, our analysis indicates that Transformers are not fully optimized on such emerging hardware, primarily due to inadequate optimizations in non-matrix computational kernels like Softmax and in heterogeneous resource utilization, particularly when processing long sequences. To address these issues, we propose an integrated approach (called GFormer) that merges sparse and linear attention mechanisms. GFormer aims to maximize the computational capabilities of the Gaudi processor's Matrix Multiplication Engine (MME) and Tensor Processing Cores (TPC) without compromising model quality. GFormer includes a windowed self-attention kernel and an efficient outer product kernel for causal linear attention, aiming to optimize LLM inference on Gaudi processors. Evaluation shows that GFormer significantly improves efficiency and model performance across various tasks on the Gaudi processor and outperforms state-of-the-art GPUs.