Greedy+Singleton is the textbook heuristic for monotone submodular maximization under a knapsack constraint: it returns the better of the cost-benefit greedy set and the best single item. Its worst-case approximation ratio, known only to lie in $[0.427, 0.42945]$ (Feldman-Nutov-Shoham; Kulik-Schwartz-Shachnai), is exactly an explicit constant $β^\star\approx 0.42944$. In the extremal instances, many cheap "filler" items wear down the value of two large optimal items while greedy spends the budget, until neither fits. This picture gives a sharper factor-revealing relaxation; we bound it analytically near the extremal parameters, up to finitely many interval-arithmetic checks, and certify it elsewhere by an exact branch-and-bound. The matching bad instances are tie-free and checkable by hand. For the version that stops at the first item that does not fit, Wolsey's bound $1-e^{-β_W}\approx 0.35780$, where $e^{β_W}=2-β_W$, is tight; Wolsey stated this for his variant without giving the instances, and we give an explicit family with no ties that also covers comparison with the best feasible singleton. The CELF rule also has ratio exactly $β^\star$.
Tokenisers such as Byte-Pair Encoding (BPE) shorten text by repeatedly merging an adjacent pair of symbols into a new one; fewer tokens mean less computation. We ask whether an efficient algorithm can choose $k$ merges that leave a string of length $n$ almost as short as possible; doing nothing is already within a factor $n$. We prove that, unless P = NP, no polynomial-time algorithm comes within a factor $c\,n/\log n$, for some constant $c>0$, already for one binary string and for BPE-style merges that replace every occurrence. The key idea is that $k$ merges form a grammar with $k$ rules: a vertex-cover gap becomes a grammar-size gap, which we amplify with relabelled copies, encode in binary and realise with explicit merges. For merges that may skip occurrences, a simple algorithm achieves $O(n/\log\log n)$, and no Karp reduction can prove hardness beyond order $n/\log n$ unless coNP $\subseteq$ NP/poly. Over a one-letter alphabet, both tokenisation problems left open by Kastreva et al. are strongly NP-complete on datasets with repeated strings.
A tokenizer's vocabulary is chosen once and used on every input, so it matters how much compression a greedy vocabulary can lose. GreedTok (Lim et al., 2025) picks $k$ tokens one at a time, each time the string that saves the most tokens; its worst-case ratio was open. We prove that GreedTok saves at least 0.3287 of the optimal number of tokens saved, on every corpus and budget, while junction corpora show that no constant above 0.3387 is possible. The lower-bound proof is written out in full and uses no computer certificates. Its key idea is to track the order of greedy's choices: given that order, the pattern greedy leaves inside an optimal token depends on two numbers, and an induction peels off one chosen string at a time; without the order, a needed inequality fails for a 20-character token. Over 256 symbols, GreedTok's ratio exceeds 0.33861 for every $k\ge 104{,}910$, and over $k^c$ symbols with $c<1/2$ it is at least 0.4386 in the limit. We also show that the objective is APX-hard: it has no PTAS unless P = NP.
Byte-Pair Encoding (BPE) builds most language-model tokenisers by repeatedly merging a most frequent adjacent pair of symbols. How much shortening can this greedy rule lose against the best $k$ merges? Kozma and Voderholzer proved that greedy achieves at least one third of the optimal shortening; the best known upper bound is $(1-e^{-2})/2\approx 0.432$. We show that one third is a barrier for packing-based analyses, which see the optimum only through a pair packing, and we prove that greedy achieves at least $0.391$ of the optimum on every input, with any tie-breaking. The key idea is rule determinism: all copies of a rule of the optimum spell the same letters, so when one greedy replacement destroys three of the optimum's merge points, half a unit can be refunded on each side. A weaker bound, $0.374$, has a short proof that needs no computer; $0.391$ adds a certificate in exact arithmetic. We also show that the worst-case ratio is exactly $3/4$ for two merges and at most $7/12$ for three.