Generative AI
SearchStart here

Module deck

Evaluation and model adaptation flashcards

Reviewing needs JavaScript. You can read every card in the list below.

Loading your cards...

Back to the module21 cards in this deck

Read all 21 cards as a list

Evaluating LLM systems

  1. Front
    Why can't a public benchmark pick your model?
    Back
    Contamination, saturation and domain mismatch. Use benchmarks to shortlist; decide on your product eval.
  2. Front
    How do you validate an LLM judge before trusting it?
    Back
    Label a sample by humans, compute Cohen's kappa against the judge (not raw percent agreement), report the label balance, and re-check periodically.
  3. Front
    Why can high percent agreement between judge and humans mislead?
    Back
    When one label dominates, guessing it already agrees often. With 95% passes, chance agreement is about 0.905, so 90% observed is no better than chance. Use kappa and look at the rare class.
  4. Front
    What is the difference between a dev set and a held-out set?
    Back
    Iterate on the dev set. Use the held-out set only to gate releases. Tuning against it turns it into a second dev set and its score stops meaning anything.
  5. Front
    What should you do before choosing evaluation metrics?
    Back
    Error analysis: read 50 to 100 real failing traces, sort them into failure categories, and measure the top categories.
  6. Front
    Why is a +3 point gain on 100 cases weak evidence?
    Back
    Sampling noise: at a pass rate near 0.8, one score is uncertain by roughly 8 points either way (illustrative). Compare variants paired on the same cases, with a bootstrap interval, or add cases.

Fine-tuning vs RAG vs prompting

  1. Front
    When is fine-tuning the right choice over RAG?
    Back
    When you need consistent style, format or behaviour at scale, have labelled examples, and prompting has hit its limit. Not for injecting facts or current knowledge.
  2. Front
    Does fine-tuning fix hallucination?
    Back
    Not by itself. It does not supply correct facts, and one 2024 study found hallucination rose as new facts were absorbed. Training on correct refusals can improve abstention (R-Tuning, NAACL 2024). Facts still need grounding via RAG.
  3. Front
    What does each lever change: prompting, RAG, fine-tuning?
    Back
    Prompting changes the instructions. RAG changes the information in the prompt at answer time. Fine-tuning changes the weights, to teach a consistent pattern of behaviour.
  4. Front
    Three things fine-tuning does poorly compared with RAG?
    Back
    Keeping facts current (needs retraining), giving citations (no source passage), and per-user access control (weights are shared by every user).
  5. Front
    What must exist before you start a fine-tune?
    Back
    Labelled examples, a strong baseline (best prompt, plus RAG if facts matter), an eval with a held-out set, a reason prompting cannot meet, and a plan to retrain when the base model changes.

Fine-tuning methods: SFT, LoRA, QLoRA and DPO

  1. Front
    Explain LoRA in two sentences.
    Back
    Freeze the weights W and learn a low-rank update B x A (rank r much smaller than d), a tiny fraction of the parameters. Adapters are small, swappable and mergeable into W.
  2. Front
    RLHF vs DPO?
    Back
    RLHF trains a reward model then optimises with RL. DPO trains directly on preference pairs with a classification-style loss: no separate reward model or RL loop, though it still uses a frozen reference model.
  3. Front
    How many trainable weights does a rank-r LoRA adapter add to a d x d layer?
    Back
    2 x d x r (A is r x d, B is d x r). For d = 4096, r = 8: 65,536, about 0.39% of the layer's 16.8 million weights.
  4. Front
    What does QLoRA add to LoRA?
    Back
    The frozen base is held in 4-bit (NF4) while adapters train in higher precision, with double quantisation and paged optimisers. It cuts memory a lot, with slower steps; check quality on your task.
  5. Front
    Which data does SFT use, and which does preference tuning use?
    Back
    SFT: prompt and ideal-response pairs. Preference tuning: a prompt with a preferred and a rejected response.

Fine-tuning data, risks and shipping

  1. Front
    A tuned model looks fine offline but worse in production. First suspect?
    Back
    A prompt-format mismatch: training used a different system prompt or chat template than production. Train on the exact serving format.
  2. Front
    Name four risks of fine-tuning.
    Back
    Forgetting (general abilities regress), overfitting (memorises training examples), safety regression (weaker safety behaviour, even with benign data) and leakage (training data overlaps your evals).
  3. Front
    What do you compare a fine-tuned model against before shipping?
    Back
    The base model with the best prompt, plus RAG if facts matter. If the tune does not clearly beat that baseline, stay with the baseline.
  4. Front
    What do you need in place so a bad adapter can be reversed quickly?
    Back
    Adapters versioned together with their data and hyperparameters, a canary release, and a rollback path to the previous adapter or the base model.
  5. Front
    What is decontamination and why does it matter?
    Back
    Removing training examples that overlap your eval data. If they overlap, the scores measure memorisation and mean little.