Time-series representation learning is a key area of research for remote healthcare monitoring applications. In this work, we focus on a dataset of recordings of in-home activity from people living with Dementia. We design a representation learning method based on converting activity to text strings that can be encoded using a language model fine-tuned to transform data from the same participants within a $30$-day window to similar embeddings in the vector space. This allows for clustering and vector searching over participants and days, and the identification of activity deviations to aid with personalised delivery of care.

Large language models (LLMs) have demonstrated substantial commonsense understanding through numerous benchmark evaluations. However, their understanding of cultural commonsense remains largely unexamined. In this paper, we conduct a comprehensive examination of the capabilities and limitations of several state-of-the-art LLMs in the context of cultural commonsense tasks. Using several general and cultural commonsense benchmarks, we find that (1) LLMs have a significant discrepancy in performance when tested on culture-specific commonsense knowledge for different cultures; (2) LLMs' general commonsense capability is affected by cultural context; and (3) The language used to query the LLMs can impact their performance on cultural-related tasks. Our study points to the inherent bias in the cultural understanding of LLMs and provides insights that can help develop culturally aware language models.

Recent developments in large language models (LLMs), while offering a powerful foundation for developing natural language agents, raise safety concerns about them and the autonomous agents built upon them. Deception is one potential capability of AI agents of particular concern, which we refer to as an act or statement that misleads, hides the truth, or promotes a belief that is not true in its entirety or in part. We move away from the conventional understanding of deception through straight-out lying, making objective selfish decisions, or giving false information, as seen in previous AI safety research. We target a specific category of deception achieved through obfuscation and equivocation. We broadly explain the two types of deception by analogizing them with the rabbit-out-of-hat magic trick, where (i) the rabbit either comes out of a hidden trap door or (ii) (our focus) the audience is completely distracted to see the magician bring out the rabbit right in front of them using sleight of hand or misdirection. Our novel testbed framework displays intrinsic deception capabilities of LLM agents in a goal-driven environment when directed to be deceptive in their natural language generations in a two-agent adversarial dialogue system built upon the legislative task of "lobbying" for a bill. Along the lines of a goal-driven environment, we show developing deceptive capacity through a reinforcement learning setup, building it around the theories of language philosophy and cognitive psychology. We find that the lobbyist agent increases its deceptive capabilities by ~ 40% (relative) through subsequent reinforcement trials of adversarial interactions, and our deception detection mechanism shows a detection capability of up to 92%. Our results highlight potential issues in agent-human interaction, with agents potentially manipulating humans towards its programmed end-goal.

Autonomous agents powered by large language models (LLMs) show significant potential for achieving high autonomy in various scenarios such as software development. Recent research has shown that LLM agents can leverage past experiences to reduce errors and enhance efficiency. However, the static experience paradigm, reliant on a fixed collection of past experiences acquired heuristically, lacks iterative refinement and thus hampers agents' adaptability. In this paper, we introduce the Iterative Experience Refinement framework, enabling LLM agents to refine experiences iteratively during task execution. We propose two fundamental patterns: the successive pattern, refining based on nearest experiences within a task batch, and the cumulative pattern, acquiring experiences across all previous task batches. Augmented with our heuristic experience elimination, the method prioritizes high-quality and frequently-used experiences, effectively managing the experience space and enhancing efficiency. Extensive experiments show that while the successive pattern may yield superior results, the cumulative pattern provides more stable performance. Moreover, experience elimination facilitates achieving better performance using just 11.54% of a high-quality subset.

We present and tackle the problem of Embodied Question Answering (EQA) with Situational Queries (S-EQA) in a household environment. Unlike prior EQA work tackling simple queries that directly reference target objects and quantifiable properties pertaining them, EQA with situational queries (such as "Is the bathroom clean and dry?") is more challenging, as the agent needs to figure out not just what the target objects pertaining to the query are, but also requires a consensus on their states to be answerable. Towards this objective, we first introduce a novel Prompt-Generate-Evaluate (PGE) scheme that wraps around an LLM's output to create a dataset of unique situational queries, corresponding consensus object information, and predicted answers. PGE maintains uniqueness among the generated queries, using multiple forms of semantic similarity. We validate the generated dataset via a large scale user-study conducted on M-Turk, and introduce it as S-EQA, the first dataset tackling EQA with situational queries. Our user study establishes the authenticity of S-EQA with a high 97.26% of the generated queries being deemed answerable, given the consensus object data. Conversely, we observe a low correlation of 46.2% on the LLM-predicted answers to human-evaluated ones; indicating the LLM's poor capability in directly answering situational queries, while establishing S-EQA's usability in providing a human-validated consensus for an indirect solution. We evaluate S-EQA via Visual Question Answering (VQA) on VirtualHome, which unlike other simulators, contains several objects with modifiable states that also visually appear different upon modification -- enabling us to set a quantitative benchmark for S-EQA. To the best of our knowledge, this is the first work to introduce EQA with situational queries, and also the first to use a generative approach for query creation.

Large Language Models (LLMs) deployed on edge devices learn through fine-tuning and updating a certain portion of their parameters. Although such learning methods can be optimized to reduce resource utilization, the overall required resources remain a heavy burden on edge devices. Instead, Retrieval-Augmented Generation (RAG), a resource-efficient LLM learning method, can improve the quality of the LLM-generated content without updating model parameters. However, the RAG-based LLM may involve repetitive searches on the profile data in every user-LLM interaction. This search can lead to significant latency along with the accumulation of user data. Conventional efforts to decrease latency result in restricting the size of saved user data, thus reducing the scalability of RAG as user data continuously grows. It remains an open question: how to free RAG from the constraints of latency and scalability on edge devices? In this paper, we propose a novel framework to accelerate RAG via Computing-in-Memory (CiM) architectures. It accelerates matrix multiplications by performing in-situ computation inside the memory while avoiding the expensive data transfer between the computing unit and memory. Our framework, Robust CiM-backed RAG (RoCR), utilizing a novel contrastive learning-based training method and noise-aware training, can enable RAG to efficiently search profile data with CiM. To the best of our knowledge, this is the first work utilizing CiM to accelerate RAG.

Large Language Models (LLMs) have made great strides in areas such as language processing and computer vision. Despite the emergence of diverse techniques to improve few-shot learning capacity, current LLMs fall short in handling the languages in biology and chemistry. For example, they are struggling to capture the relationship between molecule structure and pharmacochemical properties. Consequently, the few-shot learning capacity of small-molecule drug modification remains impeded. In this work, we introduced DrugLLM, a LLM tailored for drug design. During the training process, we employed Group-based Molecular Representation (GMR) to represent molecules, arranging them in sequences that reflect modifications aimed at enhancing specific molecular properties. DrugLLM learns how to modify molecules in drug discovery by predicting the next molecule based on past modifications. Extensive computational experiments demonstrate that DrugLLM can generate new molecules with expected properties based on limited examples, presenting a powerful few-shot molecule generation capacity.

Transformer-based language models are treated as black-boxes because of their large number of parameters and complex internal interactions, which is a serious safety concern. Mechanistic Interpretability (MI) intends to reverse-engineer neural network behaviors in terms of human-understandable components. In this work, we focus on understanding how GPT-2 Small performs the task of predicting three-letter acronyms. Previous works in the MI field have focused so far on tasks that predict a single token. To the best of our knowledge, this is the first work that tries to mechanistically understand a behavior involving the prediction of multiple consecutive tokens. We discover that the prediction is performed by a circuit composed of 8 attention heads (~5% of the total heads) which we classified in three groups according to their role. We also demonstrate that these heads concentrate the acronym prediction functionality. In addition, we mechanistically interpret the most relevant heads of the circuit and find out that they use positional information which is propagated via the causal mask mechanism. We expect this work to lay the foundation for understanding more complex behaviors involving multiple-token predictions.

The efficacy of an large language model (LLM) generated text detector depends substantially on the availability of sizable training data. White-box zero-shot detectors, which require no such data, are nonetheless limited by the accessibility of the source model of the LLM-generated text. In this paper, we propose an simple but effective black-box zero-shot detection approach, predicated on the observation that human-written texts typically contain more grammatical errors than LLM-generated texts. This approach entails computing the Grammar Error Correction Score (GECScore) for the given text to distinguish between human-written and LLM-generated text. Extensive experimental results show that our method outperforms current state-of-the-art (SOTA) zero-shot and supervised methods, achieving an average AUROC of 98.7% and showing strong robustness against paraphrase and adversarial perturbation attacks.

Reversal curse (Berglund et al., 2023) refers to the phenomenon that an auto-regressive LLM that learns "A is B" during training fails to generalize to the reverse direction "B is A", and this task is also termed as "inverse search" in Allen-Zhu and Li (2023). Although some previous works propose different methods to mitigate the reversal curse, including reversing the training dataset (Guo et al., 2024; Golovneva et al., 2024) and training on different objectives such as autoregressive blank infilling (Lv et al., 2023), these methods might negatively affect the model performance on other tasks since they either alter the dataset or the model architecture. Without dataset manipulation or changing the auto-regressive nature (causal structure) of the model, the reversal curse is hard to mitigate even with ICL strategies such as chain-of-thought (Allen-Zhu and Li, 2023; Guo et al., 2024). In this paper, we aim to theoretically study why the reversal curse happens for auto-regressive LLMs. Different from previous work that studies the capacity of (transformer-based (Vaswani et al., 2017)) LLMs through the lens of expressivity (e.g., Yun et al. (2019); Pérez et al. (2021); Feng et al. (2024)), reversal curse cannot be explained by expressivity since a model can express "A is B" is also able to express "B is A". Therefore, we analyze the reversal curse via training dynamics since even if a set of parameters can express a fact in both directions, it might not be reachable through popular training algorithms (e.g., gradient descent, AdamW (Loshchilov and Hutter, 2017)) with training data only presented in one direction. We summarize our main contributions as follows: We theoretically analyze reversal curse where training or test sequences have the from "A B" or "B A" via training dynamics of (stochastic) gradient descent under two auto-regressive models: a bilinear model (Section 3) and one-layer transformers under certain assumptions similar to Tian et al. (2023a) (Section 4). The analysis of the training dynamics of both models reveals a core reason why the reversal curse happens: the weights of the autoregressive models are asymmetric, i.e., the increase of weights from the token A to token B