The ability to automatically generate accurate protocols for scientific experiments would represent a major step towards the automation of science. Large Language Models (LLMs) have impressive capabilities on a wide range of tasks, such as question answering and the generation of coherent text and code. However, LLMs can struggle with multi-step problems and long-term planning, which are crucial for designing scientific experiments. Moreover, evaluation of the accuracy of scientific protocols is challenging, because experiments can be described correctly in many different ways, require expert knowledge to evaluate, and cannot usually be executed automatically. Here we present an automatic evaluation framework for the task of planning experimental protocols, and we introduce BioProt: a dataset of biology protocols with corresponding pseudocode representations. To measure performance on generating scientific protocols, we use an LLM to convert a natural language protocol into pseudocode, and then evaluate an LLM's ability to reconstruct the pseudocode from a high-level description and a list of admissible pseudocode functions. We evaluate GPT-3 and GPT-4 on this task and explore their robustness. We externally validate the utility of pseudocode representations of text by generating accurate novel protocols using retrieved pseudocode, and we run a generated protocol successfully in our biological laboratory. Our framework is extensible to the evaluation and improvement of language model

BioPlanner: Automatic Evaluation of LLMs on Protocol Planning in Biology | VIDEO

In Psychotherapy, {maladaptive schemas}-- negative perceptions that {an 
individual has of the self, others, or the world that endure despite objective reality}--
often lead to resistance to treatments and relapse of mental health issues such as depression, anxiety, panic attacks etc. Identification of early maladaptive schemas (EMS) is thus a crucial 
step during Schema Therapy-based counseling sessions, 
where patients 
go through a detailed and lengthy EMS questionnaire. However, such an approach is not practical in "offline" counseling scenarios, such as community QA forums which are gaining popularity for people seeking mental health support. In this paper, we 
investigate both LLM (Large Language Models) and non-LLM approaches for identifying EMS labels using resources from Schema Therapy. 
Our evaluation indicates that 
recent LLMs can be effective for identifying EMS but their predictions lack explainability and are too sensitive to precise `prompts'. 
Both LLM and non-LLM methods are unable to reliably address the {null} cases, i.e. cases with no EMS labels. However, we posit that the two approaches show complementary properties and together, 
they can be used to further devise techniques
for EMS identification.

Identifying {Early Maladaptive Schemas} from Mental Health Question Texts

Software version migration and program translation are an important and costly part of the lifecycle of large codebases. Traditional machine translation relies on parallel corpora for supervised translation, which is not feasible for program translation due to a dearth of aligned data. Recent unsupervised neural machine translation techniques have overcome data limitations by included techniques such as back translation and low level compiler intermediate representations (IR). These methods face significant challenges due to the noise in code snippet alignment and the diversity of IRs respectively. In this paper we propose a novel model called Code Distillation (CoDist) whereby we capture the semantic and structural equivalence of code in a language agnostic intermediate representation. Distilled code serves as a translation pivot for any programming language, leading by construction to parallel corpora which scale to all available source code by simply applying the distillation compiler. We demonstrate that our approach achieves state-of-the-art performance on CodeXGLUE and TransCoder GeeksForGeeks translation benchmarks, with an average absolute increase of 12.7% on the TransCoder GeeksforGeeks translation benchmark compare to TransCoder-ST.

Program Translation via Code Distillation

Planning for goal-oriented dialogue often requires simulating future dialogue interactions and estimating task progress. Many approaches thus consider training neural networks to perform look-ahead search algorithms such as A* search and Monte Carlo Tree Search (MCTS). However, this training often require abundant annotated data, which creates challenges when faced with noisy annotations or low-resource settings. We introduce GDP-Zero, an approach using Open-Loop MCTS to perform goal-oriented dialogue policy planning without any model training. GDP-Zero prompts a large language model to act as a policy prior, value function, user simulator, and system model during the tree search. We evaluate GDP-Zero on the goal-oriented task PersuasionForGood, and find that its responses are preferred over ChatGPT up to 59.32% of the time, and are rated more persuasive than ChatGPT during interactive evaluations.

Prompt-Based Monte-Carlo Tree Search for Goal-oriented Dialogue Policy Planning

Relation extraction (RE) is one of the tasks behind many relevant natural language processing (NLP) applications. Exploiting the information hidden in millions of scholarly articles by leveraging NLP, specifically RE, systems could benefit studies in specialized domains, e.g. biomedicine and biodiversity. Although deep learning (DL)-based methods have shown state-of-the-art performance in many NLP tasks including RE, DL for domain-specific RE systems has been hindered by the lack of expert labeled datasets which are typically required to train such methods. In this paper, we take advantage of the zero-shot (i.e., not requiring any labeled data) capability of pattern-based methods for RE using a rule-based approach, combined
with templates for natural language inference (NLI) transformer models. We present our hybrid method for RE that exploits the advantages of both methods, i.e., interpretability of rules and transferability of transformers. Evaluated on a corpus of biodiversity literature with annotated relations, our hybrid method demonstrated
an improvement of up to 15 percentage points in recall and best performance over solely rule-based and transformer-based methods with F1-scores ranging from 89.61% to 96.75% for reproductive condition - temporal expression relations, and ranging from 85.39% to 89.90% for habitat - geographic location relations.

A Hybrid of Rule-based and Transformer-based Approaches for Relation Extraction in Biodiversity Literature

Biomass is an organic material that normally comes from plants or animals. The discovery of converting biomass wastes into graphene is a successful route for recovering sources because it is sustainable, renewable, and abundant. For both basic research and practical applications, finding a novel way to produce graphene with a high yield and excellent crystalline structure is critical. For this research, waste disposable paper-cups will be used for the formation of graphene sheets by using KOH activation while using FeCl3 as a catalyst. The chemical activation of carbon-rich materials using KOH as the activating agent is a ground-breaking technique because of its lower activation temperature and higher yield [1]. Also, the obtained graphene will have a well-defined micropore size distribution and extremely high surface area. The graphene nanostructure can be used as a filler for thermoplastic polymers since it provides polymers of several functional characteristics, which are not accessible by using other types of fillers. For this research, the extracted graphene nanoparticles will be used as a filler in polypropylene to enhance its morphological, mechanical, and thermal properties. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) tests will be used to determine the thermal properties of the graphene reinforced polymer composites. Also, X-ray diffraction (XRD), FTIR, SEM, TEM, and tensile tests will be used to determine the mechanical, microscopic, and morphological property.es of GRPC. The results obtained from this research indicates that high yield and high-quality graphene materials can be synthesized by using biomass resources rather than conventional methods and can be used as a filler in polymers that could improve the mechanical and thermal performance when dispersed uniformly.

Morphological, Mechanical, and Thermal Properties of Polypropylene Reinforced with Graphene Nanoparticles Extracted from Paper Cups.

The Center for Energy and Environmental Sustainability (CEES) was established in October 2010 with $5M in funding from NSF. CEES Phase II award from NSF started on October 1, 2019. CEES personnel include eight faculty members, 1 program coordinator, and 4 research staff members. CEES is addressing the challenging interdisciplinary problems and is paving the way for scientific and technological breakthroughs in energy engineering to meet short- and long-term national needs. The Center’s research mission is accomplished through 3 distinct research thrust areas and strong education and outreach programs. The research areas include Bio-Energy (Energy crop cultivation and gasoline/diesel-like hydrocarbon fuel production from energy crops), Offshore Wind Energy (Massive offshore power generation and risk analysis with big data), and Environmental Sustainability (Produced water treatment and environmental impact evaluation with life cycle assessment). The Center’s education and outreach activities are carried through the Energy Engineering and Research Learning Initiative (EERLI), which includes a myriad of activities that engage K-12 students through graduate students. The researchers, including our students, have published 1 book, 185 peer-reviewed publications and have given more than 220 presentations at regional, national, and international conferences. CEES faculty and staff are also supported by other leveraged funding of about $12.82 M from NSF, USDA, Tran-SET, and industrial partners. CEES has supported more than 220 students, including 73 graduate, and 145 undergraduate students. The Center also established state-of-the-art research laboratories with high-quality equipment using leveraged funds.

Center for Energy & Environmental Sustainability (CEES)

The Centers of Research Excellence in Science and Technology (CREST) program provides support to enhance the research capabilities of minority-serving institutions (MSI). Southern University and A&M College in Baton Rouge (SUBR) established its Next Generation Composites CREST Center (NextGenC3) by a unifying theme of multiscale and multifunctional composites and related technology development of advanced fiber reinforced polymer composites with capabilities to address varying technological needs and challenges in material applications. Dr. Jahan in this colloquia session will provide an introduction and overview of her projects focusing on food wastes and byproducts. Her team at SUBR provides educational and research training opportunities to traditionally underrepresented minority (URM) students in Science, Technology, Engineering, and Mathematics (STEM) disciplines by using enhanced non-structural functionalities of the bio-composites towards a potential replacement for their inorganic-based counterparts in automotive, aerospace industries, and researchable fuel cell applications.

Potential Multifunctional Composite using carbonized coconut shell particles

CREST Center for Dynamic Multiscale and Multimodal Brain Mapping over the Lifespan [D-MAP]

The relevance of indigenous Psychologies for understanding human experience. Who gets to decide?

Multimodal binding as mereological co-constituency

Much of our knowledge come from other people. In considering how argument quality and source reliability influences message persuasiveness, we conduct a comparison of the Elaboration Likelihood Model of Persuasion and the Bayesian Model of Argumentation, which are based on different assumptions. Participants were asked to judge a fictitious character’s degree of belief in a claim given evidence. To test competing predictions, we manipulate the character’s elaboration level, the argument’s quality, and the source’s reliability. The elaboration did not moderate the main effects of argument quality and source reliability, as they both were integral to the overall message strength in both high and low elaboration conditions. Bayesian predictions have better fit with the observed data, whilst ELM predictions did not align well. Overall, the BM is supported, but we discuss how this model could be further improvised while the ELM is contested


Underline digital video library

BioPlanner: Automatic Evaluation of LLMs on Protocol Planning in Biology | VIDEO

Identifying {Early Maladaptive Schemas} from Mental Health Question Texts

Program Translation via Code Distillation

Prompt-Based Monte-Carlo Tree Search for Goal-oriented Dialogue Policy Planning

A Hybrid of Rule-based and Transformer-based Approaches for Relation Extraction in Biodiversity Literature

Morphological, Mechanical, and Thermal Properties of Polypropylene Reinforced with Graphene Nanoparticles Extracted from Paper Cups.

Center for Energy & Environmental Sustainability (CEES)

Potential Multifunctional Composite using carbonized coconut shell particles

CREST Center for Dynamic Multiscale and Multimodal Brain Mapping over the Lifespan [D-MAP]

The relevance of indigenous Psychologies for understanding human experience. Who gets to decide?

Multimodal binding as mereological co-constituency

Comparing predictions from the Elaboration Likelihood Model and a Bayesian model of argumentation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES