finding
active
finding:atlas-la-grpo-achieves-51-3-on-blink-average-improving-from-baseline-22-8ATLAS LA-GRPO achieves 51.3% on BLINK average, improving from baseline 22.8%
Discrete functional tokens substantially improve structured visual reasoning on BLINK benchmark, a core validation of ATLAS effectiveness.
Source paper
extracted_fromZiyu Guo · Rain Liu · Xinyan Chen · Pheng-Ann Heng
Neighborhood — ranked by edge-count
Hypotheses (1)
hypothesis
- ATLAS hypothesis that a compact set of high-level functional tokens (Manip, Shape, Line, Arrow, Text) suffices for multi-domain visual reasoning.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Replication of non-monotonic harness-benefit pattern on a second benchmark
- Block and Grassmannian featurizers keep utilization ≈0.6–0.66 at k=16, while Group Lasso falls to ≈0.17.finding0.734Per-feature occupancy.
- Baseline MatterGen achieves 6.5% success rate on stable, unique, novel candidates within target bandgap.finding0.719Quantitative baseline establishing the performance floor for self-correcting search improvements.
- Quantifies how much of the base model's diversity VS can recover compared to baseline prompting
- Benchmark identification rate rises from 0.08 (SFT) to 0.25 (DPO); interventions barely reduce itfinding0.717Table 11.
- Establishes low-bar baseline showing personality control without intervention is poor
- Probes fire on 36.9%/31.1% (Kimi) and 34.0%/29.9% (GLM) of τ³-bench/ShoppingBench vs monitor 3.9–8.4%finding0.713Out-of-distribution firing.
- GPT-OSS-120B achieves 5.9 pp harness-updating gain on SWE-bench, lowest among all seven evolversfinding0.712Part of full evolver-side matrix demonstrating flat but variable harness-updating across models