You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

By requesting access you agree to the CC BY-NC 4.0 license terms. Commercial use requires a separate licence from IIT Delhi.

SMRITI — Software-based Cognitive Sensing

SMRITI (स्मृति — Remembered Knowledge / Tradition) estimates cognitive load and attention from student interaction patterns — no hardware required.

Part of the VIDYA IKS-integrated adaptive education platform.

Installation

pip install smriti-sensing

Or from source:

pip install -e .

Quick Start

from smriti import SmritiEngine

engine = SmritiEngine(baseline_response_ms=3000.0)

# Record student interactions
engine.record_interaction(response_time_ms=2500, message_length=45, was_correct=True)
engine.record_interaction(response_time_ms=4200, message_length=30, was_correct=False)
engine.record_interaction(response_time_ms=6100, message_length=15, was_correct=False)

# Get cognitive state
state = engine.get_cognitive_state()
print(f"Cognitive Load: {state.cognitive_load}")
print(f"Attention:      {state.attention_level}")
print(f"Confidence:     {state.confidence}")
print(f"Signals:        {state.signals_used}")

How It Works

SMRITI estimates cognitive state from behavioral interaction patterns — response timing, engagement trends, and answer accuracy. No camera, no EEG, no biometrics — just interaction timing and behavior.

API

`SmritiEngine(baseline_response_ms=3000.0)`

Create an engine with a baseline response time (ms). Adjust based on your student population.

`engine.record_interaction(response_time_ms, message_length, was_correct=None, timestamp=None)`

Record a student interaction. was_correct is optional (not all interactions have a right/wrong answer).

`engine.get_cognitive_state() -> CognitiveState`

Returns the current estimate:

@dataclass
class CognitiveState:
    cognitive_load: float      # 0.0 (idle) to 1.0 (overloaded)
    attention_level: float     # 0.0 (distracted) to 1.0 (fully engaged)
    confidence: float          # Confidence in the estimate (0-1)
    signals_used: list[str]    # Which signals contributed

Privacy and Data Handling

SMRITI processes only behavioral interaction patterns (response timing, message lengths, answer correctness). It does not collect, store, or transmit:

Personally identifiable information (PII)
Student names, IDs, or demographic data
Message content or conversation text
Biometric data of any kind

All cognitive state estimation is performed locally in-memory. No data is persisted to disk or sent to external services unless the integrating application explicitly does so. Deployments involving minors must comply with applicable data protection regulations (e.g., India's Digital Personal Data Protection Act 2023, COPPA, GDPR-K).

Patent Notice

The methods implemented in this software are the subject of Indian Complete Patent Application No. 1536IN242 — "System and Method for Non-Invasive Cognitive State Estimation Using Behavioral Interaction Patterns in Educational Software", filed March 2026 in the name of Indian Institute of Technology Delhi. Inventor: Prof. Santhosh Sivasubramani.

The source code is released under CC BY-NC 4.0 for research and educational use. The patented method may not be used in commercial products or services without a separate licence from IIT Delhi.

License

CC BY-NC 4.0 — Free for research and educational use. Commercial use requires a license from IIT Delhi. See LICENSE for details.

Citation

@software{smriti_sensing,
  title={SMRITI: Software-based Cognitive State Sensing for Adaptive Education},
  author={Sivasubramani, Santhosh},
  year={2026},
  institution={INTRINSIC Lab, RSL Foundation, IIT Delhi},
  url={https://github.com/RSL-INTRINSICLab-IIT/smriti}
}

Contact

Prof. Santhosh Sivasubramani
Director, INTRINSIC Laboratory
RSL Foundation, Centre for SeNSE, IIT Delhi
ssivasub@iitd.ac.in
https://intrinsic.iitd.ac.in

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

Other

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Evaluation results

Signals Used
self-reported

response_time, interaction_timing, message_length, correctness
Input
self-reported

Behavioral interaction patterns (timing, length, correctness)
Output
self-reported

CognitiveState(cognitive_load, attention_level, confidence, signals_used)