BTechibtechi.in
Loading Loading
Free IIT Madras BS notes, PYQs and mock tests at btechi.in

DL for CV · Vision-language: CLIP & BLIP

25 questions · 18 marks · 2 papers

Answer key

Indian Institute of Technology Madras

BS Degree Programme · Examination & Study Paper

Deep Learning for Computer Vision

Vision-language: CLIP & BLIP

Official Paper

Examination Instructions & Guidelines

Solve on notebook/paper first before checking solutions.
MCQ: Exactly 1 option is correct. Negative marking applies.
MSQ: One or more options correct. Select all to score.
NAT: Numerical response. No negative marking.
Scientific calculator allowed for all calculations.
Official answer key is sealed at the end of this sheet.

End Term · 31 Aug 2025 (FN)End Term

Q1MSQ2 Marks·Choose all correct options
Asked 2 times Key ↓
Which of the following statements about CLIP are TRUE? (Select ALL that apply)
1. ACLIP uses a contrastive loss function to align text and image embeddings.
2. BCLIP requires image labels for training.
3. CCLIP is trained to maximize cosine similarity between embeddings of matching text-image pairs.
4. DCLIP model is trained with pretext tasks such as rotation prediction and inpainting to learn effective feature representations.
5. ETo use the CLIP model for image classification, it is imperative to finetune it on downstream datasets.
Read the passage below, then answer Q2–Q12
Consider the BLIP model architecture with an input (image-text pair) as shown below: For each of the given entities, enter the appropriate number from the figure that corresponds to it:
Based on the above data, answer the given subquestions.
Q2NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q3NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q4NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q5NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q6NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q7NAT0.55 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q8NAT0.54 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q9NAT0.54 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q10NAT0.54 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q11NAT0.54 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q12NAT0.54 Marks·Write the number
Asked 22 times Key ↓
Your answer:

End Term · 31 Aug 2025 (AN)End Term

Q13MCQ2 Marks·Choose one option
Key ↓
Given two normalized embeddings from CLIP, I = [0.2, 0.5, 0.3, 0.4] for an image and T = [0.1, 0.6, 0.3, 0.7] for a text description, compute the CLIPScore(I,T).
1. A− 69%
2. B0%
3. C69%
4. D50%
Q14MSQ2 Marks·Choose all correct options
Asked 2 times Key ↓
Which of the following statements about CLIP are TRUE? (Select ALL that apply)
1. ACLIP uses a cross-entropy loss function to align text and image embeddings.
2. BCLIP requires image labels for training.
3. CCLIP is trained to maximize cosine similarity between embeddings of matching text-image pairs.
4. DCLIP model is trained with pretext tasks such as rotation prediction and inpainting to learn effective feature representations.
5. ETo use the CLIP model for image classification, it is imperative to finetune it on downstream datasets.
Read the passage below, then answer Q15–Q25
Consider the BLIP model architecture with an input (image-text pair) as shown below:
For each of the entities given, enter the appropriate number from the figure that corresponds to it:
Based on the above data, answer the given subquestions.
Q15NAT1 Mark·Write the number
Asked 22 times Key ↓
Self-Attention _______________
Your answer:
Q16NAT0.5 Marks·Write the number
Asked 22 times Key ↓
BERT Text Encoder Bi Self Attention:__________________
Your answer:
Q17NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Image-Text Matching Loss (LITM ):__________________
Your answer:
Q18NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Image Grounded Text Encoder Bi Self Attention: ___________________
Your answer:
Q19NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q20NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Image Grounded Text Encoder Cross Attention: _________________
Your answer:
Q21NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Image-Text Contrastive Loss (LITC): ____________
Your answer:
Q22NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q23NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Causal Self Attention: _____________
Your answer:
Q24NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Your answer:
Q25NAT0.5 Marks·Write the number
Asked 22 times Key ↓
Image Grounded Text Decoder Cross Attention: _____________
Your answer:

Official Key

Master Answer Key & Explanations

Official answer keys from IIT Madras BS past papers. Click any question to review.

Reveal Key

Q1A, C

Q21

Q35

Q411

Q57

Q68

Q74

Q810

Q99

Q103

Q112

Q126

Q13C

Q14C

Q151

Q165

Q1711

Q187

Q198

Q204

Q2110

Q229

Q233

Q242

Q256

Questions and answer keys are reproduced from the official IIT Madras BS papers. Topics and repeated question types are BTechi’s own grouping. BTechi is not affiliated with IIT Madras.