Abstract
This
paper
presents
a
novel
approach
to
training
evaluation
using
Language
Models
(LLM)
and
Generative
AI
(GenAI)
build
classification
model.
The
study
aims
develop
resource-efficient
solution
for
analyzing
rubrics
transcripts,
thereby
enhancing
the
assessment
of
learning
outcomes
performance.
methodology
involves
data
collection,
preprocessing,
model
fine-tuning,
prompting,
evaluation.
A
pre-trained
LLM
is
fine-tuned
on
preprocessed
allowing
it
adapt
specific
language
patterns
structures.
generates
prompts
classify
materials
based
predefined
criteria,
with
domain
expertise
incorporated
complex
rules.
Results
demonstrate
60%
reduction
in
processing
time
evaluating
transcripts
compared
manual
assessment.
implemented
has
significantly
reduced
workload
department
improved
efficiency
analysis.
Furthermore,
model's
feedback
led
targeted
improvements
content,
resulting
higher
learner
satisfaction.
innovative
application
GenAI
offers
new
perspective
leveraging
enhance
educational
processes
manner.
Frontiers in Education,
Journal Year:
2024,
Volume and Issue:
9
Published: Dec. 4, 2024
The
emergence
of
generative
AI
in
education
introduces
both
opportunities
and
challenges,
especially
student
assessment.
This
paper
explores
the
transformative
influence
on
assessment
practices,
drawing
from
recent
training
workshops
conducted
with
educators
Global
South.
It
examines
how
can
enrich
traditional
approaches
by
fostering
critical
thinking,
creativity,
collaboration.
innovative
frameworks,
such
as
AI-resistant
assessments
Process-Product
Assessment
Approach,
which
emphasize
evaluating
not
only
final
product
but
also
student’s
interaction
tools
throughout
their
learning
journey.
Additionally,
it
provides
practical
strategies
for
integrating
into
assessments,
underscoring
ethical
use
preservation
academic
integrity.
Addressing
complexities
adoption,
including
concerns
around
misconduct,
this
equips
to
navigate
intricacies
human-AI
collaboration
settings.
Finally,
discusses
significance
institutional
policies
guiding
offers
recommendations
faculty
development
align
evolving
educational
landscape.
Higher Education for the Future,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Nov. 12, 2024
This
research
investigates
the
critical
need
to
integrate
affective
and
psychomotor
domains
alongside
cognitive
development
in
educational
systems
achieve
comprehensive
‘Exit
Outcomes’
of
Outcome-Based
Education
(OBE)
align
with
National
Higher
Qualification
Framework
(NHEQF)
descriptors.
Traditional
approaches
are
inadequate
for
these
goals,
prompting
introduction
AI-Charya
framework—a
novel,
artificial
intelligence
(AI)-driven
pedagogical
model.
Utilizing
a
qualitative
approach,
this
study
explores
limitations
existing
models
transformative
potential
generative
AI.
The
framework
provides
adaptive,
multimodal
strategies
that
personalize
learning
significantly
enhance
creative
thinking
skills.
Findings
indicate
students
engaged
show
marked
improvements
areas,
positioning
them
success
an
increasingly
automated
global
workforce.
However,
study’s
generalizability
is
limited
by
its
specific
contexts,
further
needed
assess
long-term
outcomes.
Despite
limitations,
offers
pioneering
blueprint
aligning
practices
OBE
NHEQF
standards,
equipping
holistic
competencies
required
dynamic,
future-oriented
careers.
has
significant
implications
policymakers,
educators
curriculum
developers
aiming
excellence
through
innovative
methodologies.
Advances in educational technologies and instructional design book series,
Journal Year:
2024,
Volume and Issue:
unknown, P. 379 - 404
Published: Aug. 27, 2024
The
potential
of
generative
AI
is
evident
in
every
field
life
and
education
also
experiencing
a
paradigm
shift.
Generative
opening
new
ways
assessment
the
tools
that
can
create
engaging
innovative
contents.
Through
promotion
adaptation
customization,
positioned
to
bring
about
significant
transformation
educational
process.
This
chapter
sheds
light
on
significance
higher
education.
It
offers
valuable
insights
into
possibilities
for
change
improvement
evaluations,
indicating
its
capacity
revolutionize
future
Abstract
This
paper
presents
a
novel
approach
to
training
evaluation
using
Language
Models
(LLM)
and
Generative
AI
(GenAI)
build
classification
model.
The
study
aims
develop
resource-efficient
solution
for
analyzing
rubrics
transcripts,
thereby
enhancing
the
assessment
of
learning
outcomes
performance.
methodology
involves
data
collection,
preprocessing,
model
fine-tuning,
prompting,
evaluation.
A
pre-trained
LLM
is
fine-tuned
on
preprocessed
allowing
it
adapt
specific
language
patterns
structures.
generates
prompts
classify
materials
based
predefined
criteria,
with
domain
expertise
incorporated
complex
rules.
Results
demonstrate
60%
reduction
in
processing
time
evaluating
transcripts
compared
manual
assessment.
implemented
has
significantly
reduced
workload
department
improved
efficiency
analysis.
Furthermore,
model's
feedback
led
targeted
improvements
content,
resulting
higher
learner
satisfaction.
innovative
application
GenAI
offers
new
perspective
leveraging
enhance
educational
processes
manner.
