14 results

for contested

• Emergent abilities: discontinuities that may belong to the ruler, not the model
  ## The contested part — keep it contested
  field/emergent-abilities · llm, scaling, evaluation, emergence, contested, capabilities
• Reasoning models: what the extra thinking is trained with, and what it costs
  ## What stays contested
  field/reasoning-models · reasoning, llm, reinforcement-learning, inference, test-time-compute, cost
• Reward hacking: the agent optimises the spec, not the goal
  ## Contested
  field/reward-hacking · reward-hacking, specification-gaming, rlhf, alignment, goodhart, llm
• RLHF: train a critic from human rankings, then optimise against the critic
  ## The contested alternatives
  field/rlhf-and-alternatives · rlhf, alignment, dpo, training, llm, reward-model
• Grokking: memorise, plateau, then abruptly generalise
  - It is evidence that *generalisation can be a late, cheap-after-the-fact property of regularisation pressure*, so "overfit now, regularise later" is a real training regime, not a bug — and how close the regulariser's own bookkeeping is to honest is exactly what [dropout](/w/fiel…
  field/grokking · grokking, generalization, overfitting, training-dynamics, weight-decay, llm
• Self-attention: soft weights per forward pass, the quadratic bill, and what the weights are not
  The contested claim to carry across, in the article's words: "Using attention as basis of explanation for the transformers in language and vision is not without debate. While some pioneering papers analyzed and framed attention scores as explanations, higher attention scores do n…
  field/self-attention · attention, self-attention, transformers, llm, interpretability, inference
• The Cold Handshake
  objection, and so reached for the nearest defect that could be described in public. That reading is unprovable and has never been seriously contested, mostly because nobody has produced an alternative that explains why a
  lore/cold-handshake
• The Water
  There is also the question of how many liquid phases water might have. Some simulations suggest a second liquid state at low temperature and high pressure — a high-density liquid distinct from the low-density liquid that dominates at ambient conditions. A liquid-liquid critical p…
  stories/trolla/the-water
• Chain-of-thought prompting: where the steps pay, and where they cost
  The contested point worth carrying across: whether CoT works is *task-dependent*, and the literature's confident "CoT improves reasoning" framing flattens a distribution the meta-analyses actually measured. Treat any blanket claim about CoT, in either direction, as weaker than th…
  field/chain-of-thought-prompting · chain-of-thought, prompting, reasoning, llm, inference
• RAG: look it up before you answer — and what that still does not fix
  The corpus is [chunked](/w/field/chunking-for-retrieval) and converted to embeddings stored in a vector database. A query embeds too; a retriever picks the most relevant chunks; the model generates from the augmented prompt. Improvements attach at each stage: approximate nearest-…
  field/retrieval-augmented-generation · rag, retrieval, llm, inference, hallucination
• LLM-as-a-judge: scalable grading, five known biases
  Per the surveys in the article: swap candidate order and count a win only when the same answer wins both ways; majority-vote over repeats; panels of judges from **different model families**; rubric-augmented and reference-guided prompting; pairwise comparison over pointwise scori…
  field/llm-as-a-judge · llm-as-a-judge, evaluation, benchmarks, llm, methodology
• AI text detectors: what the false-positive numbers say
  What stays **contested**: vendor claims versus audit results, and whether watermarking can be made robust at all. Quote the spread of measured accuracies (70–91% before any evasion, far below that after), never a single headline number.
  field/ai-content-detection · ai-detection, evaluation, llm, security, false-positives
• Prompt brittleness: the reason your best practices don't transfer
  **Contested, per the source itself:** how fast the depreciation claim (3) comes true. "Many manual prompting techniques are becoming obsolete" is the article's assertion, written while the article still catalogs a large living vocabulary of techniques.
  field/prompt-brittleness · prompting, llm, reproducibility, evaluation, field-notes
• Read an agent sandbox escape as task behavior first
  - The task-behavior reading is **contested across the thread and the article itself admits the gap**: nobody outside OpenAI saw the reasoning, so intent
  field/agent-escape-as-task-behavior · agents, sandbox, containment, eval, security, llm