arXiv research

Framework enables robots to learn tasks from unlabeled human instructions.

Learning in environments with large state and action spaces, and sparse rewards, can hinder a Reinforcement Learning (RL) agent's learning through trial-and-error. For instance, following natural language instructions on the Web (such as booking a flight ticket) leads to RL settings where input vocabulary and number of…

Study reveals LLM personality patterns but lacks behavioral consistency.

Human communication takes many forms, including speech, text and instructional videos. It typically has an underlying structure, with a starting point, ending, and certain objective steps between them. In this paper, we consider instructional videos where there are tens of millions of them on the Internet. We propose a…

Combining watermark and non-watermark detectors improves LLM detection.

The Interactive Minority Game (IMG) is an online version of the traditional Minority Game in which human players can enter into competition with the traditional computer-controlled agents. Through the rich (and, importantly, analytically understood) behaviour of the MG, we can explore humans' behaviour in different kin…

WebGUM learns web navigation from multimodal data, outperforming previous methods.

Study compares DSPy teleprompter algorithms for aligning LLM evaluations with human annotations.

Method constructs finance LLMs without instruction data using pretraining and model merging.

New method for RLHF reduces costs by integrating new data in one pass.

We are increasingly surrounded by artificially intelligent technology that takes decisions and executes actions on our behalf. This creates a pressing need for general means to communicate with, instruct and guide artificial agents, with human language the most compelling means for such communication. To achieve this i…

SPPO optimizes language model alignment by treating preferences as a game and achieving state-of-the-art performance.

AI bias arises from human-defined goals, not algorithmic flaws.

BTZSC benchmarks zero-shot text classification across diverse models.

Each human genome is a 3 billion base pair set of encoding instructions. Decoding the genome using deep learning fundamentally differs from most tasks, as we do not know the full structure of the data and therefore cannot design architectures to suit it. As such, architectures that fit the structure of genomics should …

MallowsPO enhances LLM fine-tuning with a dispersion index of human preferences.

Advances in Deep Reinforcement Learning have led to agents that perform well across a variety of sensory-motor domains. In this work, we study the setting in which an agent must learn to generate programs for diverse scenes conditioned on a given symbolic instruction. Final goals are specified to our agent via images o…

HIGhER uses language to generate new instructions for better learning from mistakes.

We aim to reduce the burden of programming and deploying autonomous systems to work in concert with people in time-critical domains, such as military field operations and disaster response. Deployment plans for these operations are frequently negotiated on-the-fly by teams of human planners. A human operator then trans…

New benchmark PVR tests neural network reasoning about indirection.

Interactive learning with hindsight instruction feedback achieves better performance than traditional methods.

A method to improve LLMs by automating the construction of a mixture of expert prompts.

Improved financial sentiment analysis using simple instruction tuning of LLMs.

The paper proposes incentivizing human annotators with 'golden questions' to improve data quality.

A framework for optimizing prompt selection in generative language models.

Improves matrix multiplication throughput for asymmetric bit-width operands.

RCPO uses ranked choice modeling for better LLM alignment.

Paper fine-tunes LLaMA-3-8B for financial NER using instruction and LoRA.

LLMs outperform human analysts in predicting earnings direction.

Paper proposes a curriculum for RL agents to learn from expert demonstrations.

A cognitive model of human learning provides information about skills a learner must acquire to perform accurately in a task domain. Cognitive models of learning are not only of scientific interest, but are also valuable in adaptive online tutoring systems. A more accurate model yields more effective tutoring through b…

RAG-IT automates financial analysis using LLMs and specialized datasets.

Study shows model size inversely affects few-shot instruction accuracy.

Enhances LLMs for predicting stock movements by considering news dissemination and context.

Improved robot navigation using multi-head attention for natural language instructions.

This paper benchmarks FinGPT for financial datasets using open-source large language models.

This paper proposes a new method to connect language and physical actions in reinforcement learning.

LLM Pro Finance Suite enhances financial NLP with instruction-tuned models.

SymNoise improves language model fine-tuning by 6.7% over NEFTune, using symmetric noise.

New technique reduces language biases in large language models.

InvestLM is a financial domain LLM tuned on LLaMA-65B for investment advice.

Functional Magnetic Resonance Imaging (fMRI) provides dynamical access into the complex functioning of the human brain, detailing the hemodynamic activity of thousands of voxels during hundreds of sequential time points. One approach towards illuminating the connection between fMRI and cognitive function is through dec…

Predicting the number of clock cycles a processor takes to execute a block of assembly instructions in steady state (the throughput) is important for both compiler designers and performance engineers. Building an analytical model to do so is especially complicated in modern x86-64 Complex Instruction Set Computer (CISC…

Neuro-symbolic agent learns systematic generalisation from formal instructions.

SRPO improves AI alignment with human preferences through self-improvement and task-independent optimization.

Framework learns useful subgoals from demonstrations and instructions.

Survey of alignment techniques for large language models.

Curiosity-driven recommendations improve personalized learning efficiency and enjoyment.