Search Results for author:
Found 24 papers, 11 papers with code
RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs
A promising approach is reinforcement learning from human feedback (RLHF), which leverages human feedback to update the model in accordance with human preferences and mitigate issues like toxicity and hallucinations.
GEO: Generative Engine Optimization
We facilitate systematic evaluation in this new paradigm by introducing GEO-bench, a benchmark of diverse user queries across multiple domains, coupled with sources required to answer these queries.
Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs
Our experiments with ChatGPT-3. 5 show that this bias is ubiquitous - 80% of our personas demonstrate bias; it is significant - some datasets show performance drops of 70%+; and can be especially harmful for certain groups - some personas suffer statistically significant drops on 80%+ of the datasets.
QualEval: Qualitative Evaluation for Model Improvement
In this work, we address the shortcomings of quantitative metrics by proposing QualEval, which augments quantitative scalar metrics with automated qualitative evaluation as a vehicle for model improvement.
Estimating Numbers without Regression
Despite recent successes in language models, their ability to represent numbers is insufficient.
Distraction-free Embeddings for Robust VQA
The generation of effective latent representations and their subsequent refinement to incorporate precise information is an essential prerequisite for Vision-Language Understanding (VLU) tasks such as Video Question Answering (VQA).
Exploiting Generalization in Offline Reinforcement Learning via Unseen State Augmentations
Offline reinforcement learning (RL) methods strike a balance between exploration and exploitation by conservative value estimation -- penalizing values of unseen states and actions.
C-STS: Conditional Semantic Textual Similarity
Semantic textual similarity (STS), a cornerstone task in NLP, measures the degree of similarity between a pair of sentences, and has broad application in fields such as information retrieval and natural language understanding.
Anthropomorphization of AI: Opportunities and Risks
With widespread adoption of AI systems, and the push from stakeholders to make it human-like through alignment techniques, human voice, and pictorial avatars, the tendency for users to anthropomorphize it increases significantly.
RL4F: Generating Natural Language Feedback with Reinforcement Learning for Repairing Model Outputs
Despite their unprecedented success, even the largest language models make mistakes.
ProKnow: Process Knowledge for Safety Constrained and Explainable Question Generation for Mental Health Diagnostic Assistance
LMs augmented with ProKnow guided method generated 89% safer questions in the depression and anxiety domain.
Toxicity in ChatGPT: Analyzing Persona-assigned Language Models
Large language models (LLMs) have shown incredible capabilities and transcended the natural language processing (NLP) community, with adoption throughout many services like healthcare, therapy, education, and customer service.
SPARTAN: Sparse Hierarchical Memory for Parameter-Efficient Transformers
Fine-tuning pre-trained language models (PLMs) achieves impressive performance on a range of downstream tasks, and their sizes have consequently been getting bigger.
Lila: A Unified Benchmark for Mathematical Reasoning
Mathematical reasoning skills are essential for general-purpose intelligent systems to perform tasks from grocery shopping to climate modeling.
Ranked #1 on
Mathematical Reasoning
on Lila (OOD)
Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
However, it is unknown if the models can handle more complex problems that involve math reasoning over heterogeneous information, such as tabular data.
Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering
We further design language models to learn to generate lectures and explanations as the chain of thought (CoT) to mimic the multi-hop reasoning process when answering ScienceQA questions.
Ranked #5 on
Science Question Answering
on ScienceQA
NumGLUE: A Suite of Fundamental yet Challenging Mathematical Reasoning Tasks
Given the ubiquitous nature of numbers in text, reasoning with numbers to perform simple calculations is an important skill of AI systems.
How Much Coffee Was Consumed During EMNLP 2019? Fermi Problems: A New Reasoning Challenge for AI
FPs are commonly used in quizzes and interviews to bring out and evaluate the creative reasoning abilities of humans.
Model-Advantage and Value-Aware Models for Model-Based Reinforcement Learning: Bridging the Gap in Theory and Practice
This work shows that value-aware model learning, known for its numerous theoretical benefits, is also practically viable for solving challenging continuous control tasks in prevalent model-based reinforcement learning algorithms.
Programming Puzzles
The dataset is comprehensive in that it spans problems of a range of difficulties and domains, ranging from trivial string manipulation problems, to classic programming puzzles (e. g., Tower of Hanoi), to interview/competitive-programming problems (e. g., dynamic programming), to longstanding open problems in algorithms and mathematics (e. g., factoring).
Bridging Worlds in Reinforcement Learning with Model-Advantage
Despite the breakthroughs achieved by Reinforcement Learning (RL) in recent years, RL agents often fail to perform well in unseen environments.
ADAPTIVE GENERATION OF PROGRAMMING PUZZLES
Puzzles are objective in that one can easily test the correctness of a given solution x by seeing whether it satisfies f, unlike the most common representations for program synthesis: given input-output pairs or an English problem description, the correctness of a given solution is not determined and is debatable.
Learn from Your Neighbor: Learning Multi-modal Mappings from Sparse Annotations
Many structured prediction problems (particularly in vision and language domains) are ambiguous, with multiple outputs being correct for an input - e. g. there are many ways of describing an image, multiple ways of translating a sentence; however, exhaustively annotating the applicability of all possible outputs is intractable due to exponentially large output spaces (e. g. all English sentences).
Neural-Guided Deductive Search for Real-Time Program Synthesis from Examples
In this work, we propose Neural Guided Deductive Search (NGDS), a hybrid synthesis technique that combines the best of both symbolic logic techniques and statistical models.
