We study whether Large Language Models (LLMs) latently perform multi-hop reasoning with complex prompts such as "The mother of the singer of 'Superstition' is".We look for evidence of a latent reasoning pathway where an LLM (1) latently identifies "the singer of 'Superstition"' as Stevie Wonder, the bridge entity, and (2) uses its knowledge of Stevie Wonder's mother to complete the prompt.We analyze these two hops individually and consider their co-occurrence as indicative of latent multi-hop reasoning.For the first hop, we test if changing the prompt to indirectly mention the bridge entity instead of any other entity increases the LLM's internal recall of the bridge entity.For the second hop, we test if increasing this recall causes the LLM to better utilize what it knows about the bridge entity.We find strong evidence of latent multi-hop reasoning for the prompts of certain relation types, with the reasoning pathway used in more than 80% of the prompts.However, the utilization is highly contextual, varying across different types of prompts.Also, on average, the evidence for the second hop and the full multi-hop traversal is rather moderate and only substantial for the first hop.Moreover, we find a clear scaling trend with increasing model size for the first hop of reasoning but not for the second hop.Our experimental findings suggest potential challenges and opportunities for future development and applications of LLMs.

Knowledge and language understanding of models evaluated through question answering (QA) has been usually studied on static snapshots of knowledge, like Wikipedia. However, our world is dynamic, evolves over time, and our models’ knowledge becomes outdated. To study how semi-parametric QA models and their underlying parametric language models (LMs) adapt to evolving knowledge, we construct a new large-scale dataset, StreamingQA, with human written and generated questions asked on a given date, to be answered from 14 years of time-stamped news articles. We evaluate our models quarterly as they read new articles not seen in pre-training. We show that parametric models can be updated without full retraining, while avoiding catastrophic forgetting. For semi-parametric models, adding new articles into the search space allows for rapid adaptation, however, models with an outdated underlying LM under-perform those with a retrained LM. For questions about higher-frequency named entities, parametric updates are particularly beneficial. In our dynamic world, the StreamingQA dataset enables a more realistic evaluation of QA models, and our experiments highlight several promising directions for future research.

Our world is open-ended, non-stationary, and constantly evolving; thus what we talk about and how we talk about it change over time. This inherent dynamic nature of language contrasts with the current static language modelling paradigm, which trains and evaluates models on utterances from overlapping time periods. Despite impressive recent progress, we demonstrate that Transformer-XL language models perform worse in the realistic setup of predicting future utterances from beyond their training period, and that model performance becomes increasingly worse with time. We find that, while increasing model size alone—a key driver behind recent progress—does not solve this problem, having models that continually update their knowledge with new information can indeed mitigate this performance degradation over time. Hence, given the compilation of ever-larger language modelling datasets, combined with the growing list of language-model-based NLP applications that require up-to-date factual knowledge about the world, we argue that now is the right time to rethink the static way in which we currently train and evaluate our language models, and develop adaptive language models that can remain up-to-date with respect to our ever-changing and non-stationary world. We publicly release our dynamic, streaming language modelling benchmarks for WMT and arXiv to facilitate language model evaluation that takes temporal dynamics into account.

The 2011 HRI Pioneers Workshop will be conducted in conjunction with the 2011 ACM/IEEE International Conference on Human-Robot Interaction (HRI). The 2011 HRI Pioneers Workshop will provide a forum for graduate students and postdocs to learn about the current state of HRI, to present their work and to network with one another and with select senior researchers in a setting that is less formal and more interactive than the main conference. Workshop participants will discuss important issues and open challenges in the field, encouraging the formation of collaborative relationships across disciplines and geographic boundaries.

This article combines programming by demonstration and adaptive control for teaching a robot to physically interact with a human in a collaborative task requiring sharing of a load by the two partners. Learning a task model allows the robot to anticipate the partner's intentions and adapt its motion according to perceived forces. As the human represents a highly complex contact environment, direct reproduction of the learned model may lead to sub-optimal results. To compensate for unmodelled uncertainties, in addition to learning we propose an adaptive control algorithm that tunes the impedance parameters, so as to ensure accurate reproduction. To facilitate the illustration of the concepts introduced in this paper and provide a systematic evaluation, we present experimental results obtained with simulation of a dyad of two planar 2-DOF robots.

We present a generic framework that combines Dynamical Systems movement control with Programming by Demon- stration (PbD) to teach a robot bimanual coordination task. The model consists of two systems: a learning system that processes data collected during the demonstration of the task to extract coordination constraints and a motor system that reproduces the movements dynamically, while satisfy- ing the coordination constraints learned by the ¯rst system. We validate the model through a series of experiments in which a robot is taught bimanual manipulatory tasks with the help of a human.