Yhprum on Warpcast

Yhprum pfp

thinking about why llms aren't considered funny. if we look at how kantian humor is driven by intentional incongruity/laughter arising from recognizing a mismatch between expectation and reality + how LLMs work it's a little clear why llms struggle to be universally funny👇

1 reply

0 recast

0 reaction

Yhprum pfp

this intentionality means humor depends on context (𝑐), expectation (𝐸), reality (𝑅), and shared knowledge (𝑆). formally, humor 𝐻 ≈ 𝑓(𝑐,𝐸,𝑅,𝑆).

1 reply

0 recast

0 reaction

Yhprum pfp

a critical challenge: no universal set 𝐻* of jokes exists that maximizes humor for all observers. mathematically, ∄𝐻* s.t. ∀ individuals 𝑖, humor(𝐻*,𝑖) ≥ humor(𝐻,𝑖), ∀𝐻.

1 reply

0 recast

0 reaction

Yhprum pfp

humor is subjective, mapping differently onto each observer’s internal mental model 𝑀. each model varies in prior experiences (𝑃), culture (𝐶), context (𝑐), so 𝑀(𝑖) = 𝑔(𝑃,𝐶,𝑐).

1 reply

0 recast

0 reaction

Yhprum pfp

large language models (llms) optimize for likelihood: maximizing probability of next token given prior context. formally, maximize Σ log 𝑝(tokenₙ|contextₙ₋₁,...,context₀).

1 reply

0 recast

0 reaction

Yhprum pfp

llms thus seek average predictability (high probability next tokens), conflicting fundamentally with humor’s goal: intentional unpredictability (low probability next tokens).

1 reply

0 recast

0 reaction

Yhprum pfp

kantian humor explicitly subverts expected outcomes (low-probability event). llms inherently struggle: their optimization biases output toward predictable rather than surprising tokens.

1 reply

0 recast

0 reaction

Yhprum pfp

further, intention is absent in llms: no true mental model 𝑀(𝑖), merely statistical patterns learned during training. hence, they simulate but never intentionally craft incongruity.

1 reply

0 recast

0 reaction

Yhprum pfp

in short, kantian humor requires intention (conscious incongruity), context sensitivity, and tailored mental models. llms lack all three… limiting their comedic potential significantly.

1 reply

0 recast

0 reaction

Yhprum pfp

some of the ideas to fix this also are hard, for example explicitly model surprise: define a reward function 𝑅 inversely proportional to token predictability, i.e., reward ∝ 1/𝑝(token|context).

1 reply

0 recast

0 reaction

Yhprum pfp

this fix fails practically because inverse predictability alone (1/𝑝) encourages randomness—not coherent incongruity. humor requires carefully structured surprise, not noise.

1 reply

0 recast

0 reaction

Yhprum pfp

maybe introduce an intention mechanism: condition model outputs explicitly on humor-intent prompts, shifting from purely predictive to generative incongruity.

1 reply

0 recast

0 reaction

Yhprum pfp

intent mechanisms remain simulated, not genuine: llms mimic intention without true semantic grounding, limiting their ability to intentionally craft nuanced incongruity.

1 reply

0 recast

0 reaction

Yhprum pfp

train specialized modules to learn individualized humor preferences (𝑀ₖ for each user 𝑘), ensuring personalization: humor optimized for 𝑀ₖ rather than global optimum.

1 reply

0 recast

0 reaction

Yhprum pfp

maybe mathematically, humor becomes optimization of ∑ humor(𝑀ₖ,joke), ∀𝑘 users, instead of a non-existent global frontier joke set.

2 replies

0 recast

0 reaction

Yhprum pfp

but like... individualized models (𝑀ₖ) suffer sparsity: insufficient humor-specific training data per user 𝑘. fine-tuning each 𝑀ₖ accurately at scale is unrealistic computationally.

0 reply

0 recast

0 reaction