Short-to-Long: Detached Features for Long-Horizon Value Learning
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
Short-to-Long (S2L) is a deep reinforcement learning method that trains four value networks with progressively longer discount horizons, transferring their hidden features upward through stop-gradient connections so that each network learns only from its own value loss.
Score
0.63total under rubric v2.0
- Significance0.61
- Reader interest0.65
- Audience breadth0.44
- Practical relevance0.51
- Substantiated0.87
- Technical depth0.94
- New information0.90
- Counts against: Promotional0.03
- Counts against: Overclaim0.24
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| Significance | 0.57 | 0.55 | 0.72 | 0.61 | 0.17 | +0.25 | +0.153 |
| Reader interest | 0.50 | 0.60 | 0.86 | 0.65 | 0.36 | +0.20 | +0.131 |
| Audience breadth | 0.42 | 0.35 | 0.55 | 0.44 | 0.20 | +0.10 | +0.044 |
| Practical relevance | 0.46 | 0.40 | 0.68 | 0.51 | 0.28 | +0.15 | +0.077 |
| Substantiated | 0.97 | 0.80 | 0.84 | 0.87 | 0.17 | +0.10 | +0.087 |
| Technical depth | 0.99 | 0.85 | 0.97 | 0.94 | 0.14 | +0.10 | +0.094 |
| New information | 0.90 | 0.80 | 1.00 | 0.90 | 0.20 | +0.10 | +0.090 |
| Promotional | 0.04 | 0.05 | 0.01 | 0.03 | 0.04 | −0.15 | −0.005 |
| Overclaim | 0.23 | 0.30 | 0.20 | 0.24 | 0.10 | −0.15 | −0.036 |
| Sum of contributions | +0.634 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.634 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN —
Recorded for context only, checked 24 Oct 2026, 08:54 UTC. It played no part in the ranking.